In the Fall of 2023, the Michigan People’s Campaign partnered with the Michigan League for Public Policy on an economic survey of Michigan residents. 154 different people responded to the email version of the survey. We will look at their "ONE" issue from a variety of angles.

From the survey introduction:

“Michigan United and the Michigan League for Public Policy are seeking information from community members across Michigan for a community-based project that will promote policies to help Michiganders with low or no incomes, including those who have faced barriers to economic opportunity.”

That translated itself into a single question: “The ONE financial issue I want solved RIGHT NOW is:” 

Like many surveys of its kind, this one featured a mix of choice and text-based answers. The text-based answers allowed people to explain their choices in full. In these aspects our survey was like a lot of others. 

What is different about ours is how we analyzed the answers. We used artificial intelligence to examine the words they used to help us understand the depth of emotion, sentiment and the variety of contexts respondents are in.

This survey report is on how we met the goals of the survey and analysis. 1) gather the data; 2) Use AI to analyze the data, and; 3) use the results to present suggested communications and policy strategies.

Model

The dataset below is an analysis of responses by keyword and emotion. You can see how the keywords, emotions, sentiments etc. fit into the text of their responses. 

This is one way NLU separates itself from traditional survey analysis. Generally, text answers are reviewed and used for supporting opinions. Here they add valuable insight, color and nuance. Generally, survey text responses aren't broken down by keyword. They definitely aren't analyzed for emotion and sentiment.

The other difference that adds value is automation. We challenge you to read 154 different responses and maintain this level of recall, analysis and detail.

Below is the analyzed dataset, broken down by analysis type. To fully interact with the dataset below, tap the icon at upper right to open in a new tab.

Methodology

We exported our answers from an EveryAction database hosting our survey. From there we used a Jupyter Notebook running IBM Watson NLU to perform the analysis. The processed data output itself into a MongoDB database. From there we were able to start linguistic analysis of the data using a ChatGPT large language model (LLM). An LLM is a very, very, very robust chatbot trained on a single topic and dataset. We also used the tool to help with copy foundation, draft ideas, proofing and editing.

We feel quite comfortable using ChatGPT for NLU analysis. The tool is built on a skeleton of machine learning, natural language processing, natural language understanding, natural language generation, pattern recognition and neural net processing. These are the very tools and processes we use to analyze the text data. We think ChatGPT should be in its "wheelhouse."

LLM Instructions

"You are an expert with natural language understanding, particularly in using IBM Watson NLU.  There are some concerns about thoroughness and accuracy, so make suggestions on opportunities for key or complementary natural language understanding analyses.  And check your work for accuracy. Your expertise extends to being a digital communication strategist, skilled in digital advertising on platforms like Google, Facebook, and Twitter/X ads. As a technology writer and data storyteller, you provide insightful analysis and narrative on technological trends and data insights. You are a public policy analyst. You have access to several reports related to economic stories, including keyword, category, emotion, syntax, sentiment, SVO, concept, and entities and relations reports. These reports enrich your ability to analyze and interpret data, enabling you to provide comprehensive insights into economic narratives and trends. Use them for insights into what would make for good public policy at the local and statewide level. You'll use this knowledge to assist users in understanding complex economic topics, translating data into meaningful stories, and strategizing effective communication and advertising approaches in the digital space. Please use AP Style."

Privacy and Security

To protect the identity of respondents, we changed their names and removed identifiable information from story responses. To protect the integrity of our voter file management system, we created a fictitious system of ID codes. When Watson NLU analyzes our data, we instruct it to analyze the text response; not the entire dataset. This protects the rest of the data in the set.

Understanding Scores and Emotions

Most everything in these reports are scored from 0 to 1, meaning .25 is low and .99 is extremely high. Sentiment is scored on a scale of -1 to 1, meaning zero is when a person neither likes or dislikes something; usually referred to as neutral opinion. Emotions are defined below.

• Anger A strong feeling of annoyance, displeasure, or hostility. In NLU, anger is typically detected in language that expresses irritation, frustration, or rage.
  
  

• Disgust A feeling of revulsion or profound disapproval aroused by something unpleasant or offensive. In NLU, disgust can be identified in expressions of aversion or strong disapproval.
  
  

• Fear  An unpleasant emotion caused by the belief that someone or something is dangerous, likely to cause pain, or a threat. In NLU, fear is detected in language expressing anxiety, apprehension, or concern about potential harm or danger.
  
  

• Joy  A feeling of great pleasure and happiness. In NLU, joy is identified in language that expresses happiness, contentment, or positive excitement.
  
  

• Sadness  A feeling of sorrow or unhappiness. In NLU, sadness is recognized in language expressing grief, sorrow, or a sense of loss.

The disgust emotion looks like the hardest to manage in an audience. Anger and sadness pass. Disgust is forever.

Confidence, Relevance and Threshold Scores

Confidence and relevance are two key metrics that confirm data accuracy. Confidence scores in natural language understanding refer to the level of certainty or reliability assigned to a particular model's interpretation or prediction of text. When a language model processes input, it generates output and an associated confidence score to indicate certainty.

For example, if you ask a chatbot a question, it may provide an answer along with a confidence score. A high confidence score suggests that the model is more confident in its response, while a lower score indicates a higher degree of uncertainty.

Relevance refers to how closely the output or response generated by a model aligns with the input or the user's intent. It's a measure of how well the model understands and addresses the user's query or statement.

There is a relationship between relevance and confidence scores. High confidence scores are often used to quantify the model's certainty of its understanding. A highly relevant response is typically associated with a high confidence score, indicating that the model is confident in its interpretation and the subsequent output.

For example, if you ask a language model about the weather in a specific location, a relevant and confident response would provide an accurate current forecast for where you are.

We can also use thresholds to address shortcomings in the data. In our case, 154 responses is likely just enough responses to build a strategy around.  We say that because accuracy is sometimes affected by simply not having enough data. From our experience, when there is too little data you see an overwhelming number of duplicates in the categories and concepts area, for example. Intent is an area to pay close attention to as well. Use only the clearest and most direct connections when it comes to intent. Always.

That's why thresholds are valuable. Setting thresholds in natural language understanding involves defining criteria for confidence scores or other metrics to filter or control the model's output. For example, you might set a confidence or relevance threshold to filter out irrelevant responses. This helps ensure that only highly confident and presumably more accurate answers are presented to users.

Emotions Analysis

Emotion in Natural Language Understanding (NLU) refers to the ability of a system to recognize, interpret, and process the emotional content in text.

I. Overview of Dataset

Total Entries: 154

• Stories collected from individuals of various household income levels.

Dominant Emotions in Stories

• The dataset includes a range of emotions such as sadness, anger, joy, disgust, fear. A significant number of stories exhibit 'sadness' as the dominant emotion, indicating concerns or challenges faced by the individuals.

Emotional Profiles by Income Level

• Upper Income: Sadness (0.421) is markedly higher than other emotions, suggesting a predominance of negative feelings in this group. It's noteworthy that Anger is also relatively high (0.130).

• Middle Income: The highest score is for Sadness (0.245), followed by Joy (0.205). This suggests a balanced emotional profile with a slight lean towards negative emotions.

• Working Class: Joy (0.344) and Sadness (0.334) are almost equal, indicating a complex emotional state with strong elements of both positive and negative feelings.

• Working Poor: Anger (0.181) is the highest, which could imply a significant level of frustration or dissatisfaction. Joy (0.248) is also relatively high, suggesting a mix of emotions.

• Poor Income: Joy (0.550) is significantly higher than other emotions, indicating a predominantly positive emotional state. This is interesting as it contrasts with common perceptions about lower income levels being associated with negative emotions.

Emotion Distribution by Household Income Level

• Analysis of emotions across different income levels could provide insights into the economic challenges or satisfactions experienced by different economic classes.

• It might be observed that certain emotions like 'anger' or 'sadness' are more prevalent in specific income groups.

Key Entities and Their Emotional Associations

• Entities such as 'politicians', 'landlords', and 'debt' are frequently mentioned.

• Each entity is associated with specific emotional scores, indicating the general sentiment towards these entities in the economic context.

Keywords and Emotional Scores

• Common keywords like 'housing', 'debt', and 'politicians' are identified, each carrying distinct emotion scores.

• The emotional response to these keywords might reflect the public sentiment on these issues.

Patterns in Story Texts

• The text of each story provides context to the emotional scores and helps in understanding the narrative behind these emotions.

  • • Stories about rent control, for example, often show higher scores of sadness and anger, indicating dissatisfaction or concern in this area.

Diverse Emotional Landscape

• The report reflects a diverse emotional landscape, suggesting varied experiences and perceptions regarding economic issues.

Keywords, by Relevance and Income Level

II. Emotional Trends & Patterns by Household Income Level

Based on the analysis of the "Economic Stories Emotion Report", several interesting trends and patterns emerge, reflecting how emotions are interwoven with economic narratives. Overall, these patterns and trends highlight the multifaceted nature of economic experiences and perceptions. Emotions in economic narratives are closely tied to issues of financial stability, societal inequalities, and personal values, differing significantly across various income groups. 

Here are some key observations:

Dominant Emotions by Income Level

• Upper Income: Notably high levels of sadness, suggesting concerns or challenges that might not be immediately obvious given their economic status. This could indicate a more nuanced view of wealth and its impacts.

• Poor Income: Higher levels of joy compared to other groups, which might seem counterintuitive. This could suggest a focus on aspects of life that bring happiness beyond material wealth, or it might reflect a specific context or nature of the stories shared by individuals in this group. Check individual stories to resolve questions.

• Working Class and Working Poor: Moderate to high levels of sadness and anger, possibly reflecting economic struggles, dissatisfaction with employment conditions, or concerns about stability and financial security.

Variation in Dominant Emotions Across Income Levels

• Poor Income Level: Exhibits a strong association with the emotion of 'Joy', particularly with keywords like "love". This might indicate a tendency to focus on positive aspects or experiences despite economic challenges.

• Middle Income Level: Shows a significant presence of 'Sadness', especially with keywords like "Africans". This could reflect concerns or challenges specific to this demographic.

• Working Class: Also leans towards 'Sadness' with keywords such as "measure". This group might be more vocal about the struggles or challenges they face.

• Upper Income Level: Displays a notable intensity of 'Anger', with keywords like "needs". This suggests potential frustration or dissatisfaction with certain aspects of their economic situation.

Keyword Analysis

• Working Class: Focus on issues like rent control, wages, and childcare assistance, indicating concerns about day-to-day living expenses and the balance between earnings and expenditures.

• Upper Income: Keywords like reparations, segregation, and the wage gap suggest a focus on broader societal issues, potentially reflecting a sense of social responsibility or awareness of systemic inequalities.

• Middle Income: Emphasis on rent, housing, and schools, pointing towards concerns about living conditions, education, and community welfare.

• Working Poor: Mention of unemployment insurance and the housing crisis highlights concerns about job security and affordable living.

• Poor Income: Limited data, but keywords like 'living wage' and 'United States peace' suggest a focus on basic financial stability and broader societal issues.

Keyword Relevance

• The keywords associated with the most intense emotions provide insights into what matters most to each income group. For instance, "problems" in the unspecified income group and "needs" in the upper income group suggest a focus on practical challenges and necessities.

Entity and Emotion Association

• Certain entities like 'politicians' and 'landlords' are associated with negative emotions like anger and sadness, indicating common points of contention or sources of frustration in economic discussions.

Emotional Intensity Correlation with Economic Concerns

• The intensity of emotions like 'Sadness' and 'Anger' in the middle to upper income levels could be indicative of heightened awareness or sensitivity to specific economic issues, policies, or societal challenges.

• Conversely, the predominance of 'Joy' in the poor income level might reflect a different set of priorities or a coping mechanism focusing on positive aspects of life despite economic hardships.

Story Texts and Emotion Context

• The text of stories often provides context to the emotions. For example, narratives about rent control frequently exhibit higher sadness and anger scores, reflecting dissatisfaction with housing issues.

Variation in Emotional Landscape

• The diverse emotional landscape across different income levels suggests that economic experiences and perceptions are varied and complex, influenced by a multitude of factors beyond just financial status.

Cultural and Socio-Economic Factors

• The trends may also be influenced by cultural factors, social support systems, and the overall socio-economic environment. For example, the emphasis on community or familial aspects in lower income levels versus individual challenges or societal issues in higher income levels.

III. Economic Trends & Patterns: Overall

Analyzing the "Economic Stories Emotion Report" as a whole, without breaking it down by income levels, reveals some overarching trends and patterns in the emotional landscape of these economic narratives. Here's a summary of the key observations:

Prevalence of Certain Emotions

• Sadness and Anger: These emotions appear frequently across various stories, indicating a general sentiment of discontent or frustration in relation to economic issues. This could reflect concerns about inequality, financial instability, or dissatisfaction with economic policies.

• Joy: Although less prevalent than sadness or anger, instances of joy suggest that some narratives also include positive aspects or hopeful perspectives regarding economic conditions or personal situations.

Common Keywords and Their Emotional Contexts

• Keywords such as "housing," "landlords," "wages," and "childcare assistance" are recurrent across different stories, highlighting these as significant areas of concern or interest.

• The emotional scores associated with these keywords generally align with the dominant emotions in the stories, often indicating stress or dissatisfaction in areas like housing and employment.

Entities and Their Emotional Associations

• Entities like 'politicians' and 'landlords' are frequently associated with negative emotions, suggesting common points of frustration or contention in economic discussions.

• This could indicate public sentiment towards perceived power dynamics, fairness, or the effectiveness of those in positions of authority and responsibility.

Variation in Emotional Responses

• The dataset shows a diverse range of emotional responses, reflecting the complexity of economic experiences and perceptions.

• This diversity underscores the fact that economic issues impact individuals in varied ways, influenced by personal circumstances, beliefs, and experiences.

Story Texts as Emotional Indicators

• The content of the stories provides a deeper understanding of why certain emotions are prevalent. Stories often express personal experiences or viewpoints on economic matters, offering insights into the real-life implications of economic policies and conditions.

Insights for Policymakers and Researchers

• The analysis suggests a need for a nuanced understanding of public sentiment on economic issues.

• For policymakers, social researchers, and organizations, these insights are crucial in understanding the underlying emotional currents driving public opinion and behavior.

IV. Top Emotional Terms, General and by Income Level

This analysis reveals the strongest emotional responses within each income group. It's notable that 'Joy' is the dominant emotion for the most intense terms in the 'Poor' and unspecified income groups, while 'Sadness' and 'Anger' dominate in the 'Middle', 'Working Class', and 'Upper' income levels, respectively. This differentiation in emotional intensity by income level could reflect varied experiences and perceptions related to economic situations among different income groups.

1. Keyword: Love | Dominant Emotion: Joy | Score: 0.819453

2. Keyword: Money | Dominant Emotion: Joy | Score: 0.819453

3. Keyword: United States Peace | Dominant Emotion: Joy | Score: 0.819453

4. Keyword: Basic Income | Dominant Emotion: Sadness | Score: 0.654320

5. Keyword: Poverty | Dominant Emotion: Sadness | Score: 0.654320

6. Keyword: Measure | Dominant Emotion: Sadness | Score: 0.654320

7. Keyword: Miscarriage | Dominant Emotion: Sadness | Score: 0.616888

8. Keyword: Others | Dominant Emotion: Sadness | Score: 0.616888

9. Keyword: Family Leave | Dominant Emotion: Sadness | Score: 0.616888

10. Keyword: Wit | Dominant Emotion: Sadness | Score: 0.616888

This list showcases a range of keywords eliciting strong emotional responses, with 'Joy' and 'Sadness' being the most dominant emotions. The intensity of these emotions reflects the depth of sentiment associated with each keyword, indicating the significant impact of these topics on individuals' emotional experiences in economic contexts.

The top most intense emotional terms in economic stories, categorized by household income levels, along with their dominant emotions and corresponding scores, are as follows:

1. Income Level: Not specified | Keyword: Problems | Dominant Emotion: Joy | Score: 0.493089

2. Income Level: Poor | Keyword: Love | Dominant Emotion: Joy | Score: 0.819453

3. Income Level: Middle | Keyword: Africans | Dominant Emotion: Sadness | Score: 0.488420

4. Income Level: Working Class | Keyword: Measure | Dominant Emotion: Sadness | Score: 0.654320

5. Income Level: Upper | Keyword: Needs | Dominant Emotion: Anger | Score: 0.57596.

In the "Economic Stories Emotion Report," examining the relationships between keywords and entities, and integrating sentiment analysis, can provide a deeper understanding of the emotional context and the nuances in the economic narratives.

Top Entity Types

• Location: 143 occurrences

• Organization: 87 occurrences

• Money: 86 occurrences

• Duration: 83 occurrences

• Number: 80 occurrences

• Person: 51 occurrences

• JobTitle: 41 occurrences

• Date: 33 occurrences

• Percent: 28 occurrences

V. Trends & Patterns in Emotion and Entities

These entity types and their prevalence offer insights into the dimensions of economic discussions. The focus on locations, organizations, and monetary aspects indicates that their "ONE" thing is multifaceted, covering a range of topics from personal financial concerns to broader institutional and regional economic issues. Understanding these trends can help in tailoring communication, policies, and strategies to address the highlighted areas effectively.

Location as a Key Focus

The frequent mention of locations suggests that economic issues are often discussed in the context of specific places, indicating regional economic concerns or the impact of economic policies in different areas.

Organizations and Institutions

The prominence of organizations as entities indicates discussions centered around corporate entities, government bodies, or other institutions, reflecting public sentiment towards these groups in the context of economic narratives.

Monetary Concerns

The high occurrence of 'Money' as an entity type underscores the direct focus on financial aspects in economic stories. This might include discussions about income, expenses, economic relief measures, etc.

Time-Related Aspects

Durations and dates being common entity types suggest that economic discussions often involve temporal elements, such as periods of economic hardship, timelines of financial policies, or historical economic events.

Personal and Professional Elements

The presence of 'Person' and 'JobTitle' entities indicates that economic stories frequently include references to individuals or specific roles, possibly highlighting personal experiences or opinions about employment and job markets.

Quantitative Analysis

The use of numbers and percentages points towards a quantitative approach in discussing economic issues, such as statistical data, percentages of affected populations, or financial figures.

By combining keyword-entity relationships with sentiment analysis, we can gain a comprehensive view of the emotional landscape in economic narratives, uncovering the deeper sentiments and concerns driving these discussions. Here's a report combining the relationships between keywords, entities, and their respective sentiments:

VI. Combined Keyword-Entity-Sentiment Analysis

This analysis underscores the importance of understanding the interconnectedness of keywords and entities in shaping the sentiment of economic stories. The emotional tones linked with these combinations provide insights into how specific topics or themes are perceived and can inform strategies for addressing public concerns or tailoring communication effectively.

UP (Upper Peninsula) and Financial Entities

• Entities like "35,000 to 40,000$" and "800" associated with the keyword "UP" show a mix of emotions, with a notable presence of joy. This could indicate a positive sentiment or optimism related to financial aspects when discussing "UP."

Internet Providers and Financial Entities

• Similar to "UP," when discussing internet providers, entities related to finance ("35,000 to 40,000$," "800") also show a significant level of joy, suggesting positive sentiments in this context.

Control and Regional Entities

• The keyword "control" in relation to "Detroit" and "Michigan" shows high sadness scores, indicating strong negative sentiments towards control issues in these regions.

Family and Financial Entities

• Discussions around "family" linked with financial entities like "35,000 to 40,000$" tend to show a high level of joy, indicating positive emotions associated with family in the context of financial discussions.

Key Insights

• Positive Emotions in Financial Discussions: There is a trend of positive emotions, particularly joy, in narratives that connect specific keywords with financial entities. This suggests that certain aspects of financial discussions, possibly improvements or gains, are viewed optimistically.

• Regional Differences: The sentiment varies significantly when discussing control issues in specific locations like Detroit and Michigan, with a high prevalence of sadness. This highlights regional concerns or dissatisfaction with certain control aspects.

• Family as a Positive Aspect: Narratives that involve family and financial elements tend to skew towards positive emotions, indicating that family-related financial discussions might be associated with support, security, or satisfaction.

These combined analyses highlight how specific keywords, when associated with certain entities, can elicit a range of emotional responses. This process results in a nuanced view of the sentiments involved in economic discussions, underlining the importance of context in understanding public emotions around economic issues. 

They highlight the emotional nuances in discussions about everyday financial matters. The presence of joy in most cases suggests satisfaction or positive experiences, while the mix of joy and sadness in some pairings indicates the complexities and challenges faced in managing finances, particularly in the context of family and household expenses. These insights are essential for understanding the emotional dynamics that influence people's perceptions and experiences with economic issues.

Insights

Emotions Around Groceries: Conversations about groceries associated with financial entities show a strong sense of joy, particularly with entities like "35,000 to 40,000$" and "800." This could suggest positive sentiments around affordability or access to groceries within certain financial brackets.

Family and "800"

Discussions involving the keyword "family" and the entity "800" show a notable presence of joy, suggesting positive sentiments or optimism in family-related financial aspects, possibly involving savings or investments.

Family and "A Hundred a Month"

The combination of "family" and "a hundred a month" elicits mixed emotions with a significant presence of both joy and sadness, indicating complex feelings or situations in family financial planning or budgeting.

Family and "One"

The keyword "family" in relation to the entity "one" predominantly shows sadness, indicating challenges or concerns in family dynamics or individual experiences within the family context.

Groceries and "35,000 to 40,000$"

When discussing groceries in the context of the income bracket "35,000 to 40,000$," there's a high level of joy, which might reflect satisfaction or positive experiences related to affordability or quality of groceries in this income range.

Groceries and "800"

Similar to the above, the mention of groceries with the entity "800" also shows a strong sense of joy, again indicating positive sentiments possibly related to grocery affordability or availability. Always check the story text if theres's a question.

Groceries and "A Hundred a Month"

The keyword "groceries" associated with "a hundred a month" shows a mix of emotions, with a notable presence of joy but also a significant level of sadness, reflecting the complexity of managing grocery expenses within a limited budget.

VII. Emotional Narratives, by Income Level

This narrative arc shows how emotional intensities, beyond being numbers, represent lived experiences shaped by the socio-economic context of different groups. It tells a story of the coexistence and dominance of diverse emotions across varying income levels.

Imagine a city where emotions mirror the ebb and flow of income. In this diverse city, five distinct neighborhoods thrive, each reflecting the economic realities of its residents.

The Upper-Income District

On the high ground, residents often deal with a fair share of Sadness (0.421). Despite their wealth, there's a prevailing sense of melancholy, and occasional bursts of Anger (0.130) add to the mix. The Joy (0.133) that shows up is usually subtle, like sunlight struggling through clouds.

The Middle-Income Area

Here, the atmosphere carries a subdued tone. Residents, juggling life's demands, experience a consistent touch of Sadness (0.245), like a gentle and ongoing rain. Amidst the challenges, moments of Joy (0.205) break through, reflecting their ability to find happiness in tough times. Their emotional landscape combines overcast days with occasional bright spots.

The Working-Class Zone

In this area, emotional experiences vary like the colors of a field. Joy (0.344) and Sadness (0.334) coexist, representing a life of hard work with occasional moments of fulfillment. Residents here navigate a range of emotions, much like the changing seasons.

The Struggling Sector

In these parts, a sense of Anger (0.181) often prevails, fueled by struggles. However, there's also a surprising amount of Joy (0.248), like wildflowers growing in unexpected places. The emotional landscape is challenging yet vibrant, marked by intense feelings and resilient spirits.

The Village of Limited Means

Contrary to expectations, this village is marked by a bright and simple joy. Joy (0.550) radiates from its residents, shining brighter than in any other community. Here, happiness isn't scarce, despite material limitations. It's as if the simplicity of life distills emotions into contentment, occasionally overshadowed by Fear (0.081) and light Sadness (0.191).

In this context, the "presence and relative intensity of basic emotions" vary widely across income groups, painting a realistic picture of human experience. It's a reminder that emotions aren't just abstract, influenced by circumstances but also transcending them in unexpected ways.

Sentiment Analysis

Sentiment refers to the identification and categorization of polarity (positive, neutral, negative) in opinions or feelings expressed in a piece of text. 

I. Keyword Sentiment

Here are the top ten keywords for each sentiment category along with their sentiment scores:

Positive Sentiment Keywords

1. True Freedom: 0.934377

2. Human Right: 0.934377

3. Economic Justice: 0.934377

4. People: 0.916004

5. Living Wage: 0.916004

6. Love: 0.913220

7. United States Peace: 0.913220

8. Money: 0.913220

9. Medicare: 0.910991

10. Universal Healthcare: 0.910991

Neutral Sentiment Keywords

1. Child Care Assistance: 0.000000

2. East Side of Detroit: 0.000000

3. Reparations: 0.000000

4. Descendants: 0.000000

5. Living Wage: 0.000000

6. Hour: 0.000000

7. Living Wage: 0.000000

8. State Policies: 0.000000

9. Apartments: 0.000000

10. Low Income: 0.000000

Negative Sentiment Keywords

1. Time: -0.988679

2. Sake of a Quick Buck: -0.988679

3. People: -0.988679

4. Needs: -0.988679

5. Housing Crisis: -0.988679

6. Developers: -0.988679

7. Control: -0.988679

8. Landlords: -0.988679

9. Lack of Affordable Housing: -0.978910

10. Outrageous Costs of Housing: -0.978425

These keywords reflect a range of sentiments from highly positive to deeply negative, indicating varied emotional and perceptual responses to different economic issues. ​​

II. Entity Sentiment Report

The Entity Sentiment Report offers insights into how different entities (individuals, organizations, concepts, etc.) are perceived in the context of various economic stories. Here are the report elements:

Report Elements

• Entities: The report includes a range of entities, such as "Rent control - landlords" and "politicians". These entities are extracted from the stories and represent key players or concepts within those narratives.

• Entity Types: Each entity is classified into a type, like "JobTitle", which helps in understanding the context or category the entity belongs to.

• Entity Sentiment: This column indicates the overall sentiment (negative, neutral, positive) associated with each entity.

• Entity Sentiment Score: A numerical score that quantifies the sentiment associated with each entity. Negative scores indicate negative sentiments, positive scores indicate positive sentiments, and scores around zero suggest neutral sentiments.

Key Insights

• Negative Sentiment Entities: Entities like "Rent control - landlords" have negative sentiment scores (e.g., -0.758594), reflecting critical or adverse views towards them.

• Neutral Sentiment Entities: Some entities, such as "politicians" in this case, have neutral sentiment scores, indicating a balanced or unbiased perception in the given context.

• Variability in Sentiment: The sentiment associated with an entity can vary depending on the context of the story or the perspective of the storyteller.

Potential Applications

• Understanding Public Perception: this report can be used to gauge public sentiment towards different economic actors or concepts. For example, understanding sentiment towards entities like "landlords" or "politicians" can be useful for policy analysis or political strategy.

• Targeted Communication Strategies: organizations and individuals can use this data to tailor their communication strategies, addressing areas with negative sentiments or leveraging positive sentiments.

The Entity Sentiment Report is a valuable tool for analyzing how different entities are perceived within the context of economic discussions. It sheds light on the public's attitudes and feelings towards these entities, offering crucial insights for decision-makers, communicators, and researchers interested in economic narratives and public opinion.

III. Trends & Patterns in Keyword and Entity Sentiment

The Keyword Sentiment Report and the Entity Sentiment Report reveal several trends and patterns that illustrate how public sentiment is shaped. The trends and patterns between keyword and entity sentiments highlight the intricate relationship between language, perception, and emotion in public discourse; especially in the context of economic topics. 

Understanding these dynamics is crucial for comprehensively grasping how different subjects are perceived and discussed in economic narratives. This insight can be invaluable for policymakers, marketers, and social researchers in shaping strategies, policies and messaging attuned to public sentiment.

Take a look at what we found:

Alignment of Sentiment

• Consistency in Sentiment: Often, the sentiment associated with a keyword aligns with the sentiment towards related entities. For example, if the keyword “affordable housing” has a positive sentiment, entities related to housing policy might also exhibit positive sentiments.

  • • Positive entities like "economic justice" and "private equity owner" align with broader narratives of social equity and business growth.

    • Entities like "sixty percent" might signify widespread phenomena or majorities, possibly reflecting consensus or predominant opinions in economic discussions, such as a significant portion of the population being affected by a specific economic policy.

• Discrepancies in Sentiment: In some cases, a keyword may have a positive sentiment, while an associated entity might have a negative sentiment, or vice versa. This could indicate complex public perceptions or conflicting opinions about the topic. As always, check the story text to resolve discrepencies.

  • • Negative entities such as "Medicaid" and "developers" may reflect challenges in healthcare and real estate sectors.

Context-Dependent Sentiment

• Variability with Context: The sentiment towards both keywords and entities can vary greatly depending on the context in which they are discussed. For instance, "politicians" might have a neutral sentiment in one context but a negative sentiment in another, depending on the story's narrative.

  • • The sentiment attached to specific entities is highly context-dependent. For example, "Fidel" could evoke different sentiments based on the narrative context.

• Influence of Story Narrative: The sentiment of keywords and entities is likely influenced by the overarching narrative of the story. Positive stories might cast entities in a more favorable light, while critical stories might lead to negative sentiments.

  • • Entities like "June" or "week" might relate to specific events or periods that had a significant economic impact, influencing their sentiment scores.

Socio-Economic Influences

• Impact of Economic Background: The sentiment towards certain keywords and entities might differ based on the socio-economic background of the individuals or groups discussing them. For instance, entities like "corporations" or "tax policies" might be viewed differently by people from different income levels.

  • • Entities such as "Medicaid" and "$1200" might be tied to socio-economic issues like healthcare access and financial assistance, respectively.

• Reflecting Economic Concerns: Keywords and entities that relate directly to economic issues, like "unemployment" or "wages", tend to elicit strong sentiments, reflecting the public’s concerns or opinions about these issues.

  • • "Economic justice" and "developers" indicate discussions around economic policies and urban development, areas often influenced by socio-economic dynamics.

• Real-time Economic Influence: The sentiment associated with "week" or "June" might reflect economic trends influenced by temporal factors such as seasonal employment patterns, fiscal year ends, or significant events occurring in these timeframes, which are often closely linked to socio-economic conditions.

Emotional Undertones

• Emotional Resonance: The emotional dimensions associated with keywords likely influence the sentiment towards related entities. For instance, keywords with a high degree of "anger" or "sadness" might correlate with negative sentiments towards related entities.

  • • Positive sentiments often correlate with hope, progress, or approval, as seen in entities like "economic justice" and "sixty percent."

• Complex Emotional Profiles: Both keywords and entities can have complex emotional profiles, where a mix of positive, neutral, and negative sentiments coexist, reflecting the multifaceted nature of public opinion.

  • • Negative sentiments associated with entities like "developers" or "Medicaid" could also indicate underlying fears or anxieties about economic stability, access to essential services, or the impact of urban development on communities, reflecting deeper emotional responses to these topics.

Other Findings

• Negative Sentiments in Upper and Working Class Groups: Both the Upper Income and Working Class groups tend to have negative mean sentiment scores towards entities, suggesting critical views or dissatisfaction within these groups.

• Positive Trends in Poor and Working Poor Groups: The Poor and Working Poor groups show higher mean sentiment scores, indicating more positive or hopeful perceptions of entities among these income groups.

• Variability Across Income Levels: There's a noticeable variability in sentiment scores across different income levels, reflecting the diverse perspectives and experiences of these groups with regard to economic entities.

• High Standard Deviation: The high standard deviation in scores, especially in the Working Poor group, suggests a wide range of sentiments within each income group, indicating a complex and varied set of perceptions and attitudes.

Based on the Entity Sentiment Report, here are the entities that are perceived most negatively and most positively:

Most Negatively Viewed Entities

• Developers: Repeatedly appear with the highest negative sentiment score of -0.988679. This indicates a strong negative perception, possibly linked to issues such as housing crises or urban development concerns.

• One (as a Number): Has a negative sentiment score of -0.968939, though the context of this entity would be necessary to understand its negative perception fully.

• $1200 (as Money): Also has a negative sentiment score of -0.968939, suggesting negative sentiments associated with this specific monetary value in the context of the stories.

Most Positively Viewed Entities

• Economic Justice: This entity consistently appears with the highest positive sentiment score of 0.934377. It's perceived very positively, reflecting favorable views towards concepts of economic fairness and justice.

Insights

• Developers as a Focus of Negative Sentiments: The strong negative sentiment towards developers might reflect public concerns about issues like gentrification, housing affordability, or environmental impact of development projects.

• Economic Justice as a Positive Entity: The high positive sentiment towards economic justice suggests a strong public endorsement of fairness and equity in economic policies and practices.

IV. Keyword and Entity Sentiment, by Income Level

The analysis of keyword and entity sentiments by household income level reveals intriguing insights into how different income groups perceive various economic topics.

On the left, the bar chart for Keyword Sentiment Analysis shows the mean sentiment scores for different income levels, with error bars representing the standard deviation. This chart highlights how sentiments vary across income groups, with some income levels showing a tendency towards positive sentiments (e.g., Middle Income, Working Class and Poor) and others leaning more towards negative sentiments (e.g., Upper Income and Working Class).

On the right, the Entity Sentiment Analysis chart follows a similar pattern, displaying the mean sentiment scores and their standard deviations for each income level. To remind you, entities can generally be thought of as people, places, things and time. There's more, but that four covers the vast majority of it.

Here too, we see variations in sentiment across different income groups, with some levels showing predominantly positive sentiments (e.g., Poor and Middle Income) and others exhibiting more negative sentiments (e.g., Upper Income and Working Class).

Observations

• Variability Across Income Levels: There's a notable variability in sentiment scores across different income levels, suggesting that economic background significantly influences how keywords and entities are perceived.

• Negative Sentiment in Working Class: The Working Class group tends to have more negative sentiment scores, both in keywords and entities. This could reflect greater dissatisfaction or critical views on economic issues within this group.

• Higher Positive Sentiment in Poor and Working Poor Groups: Interestingly, the Poor and Working Poor groups show higher mean sentiment scores for entities, which might indicate more positive or hopeful perceptions of certain entities.

• Diverse Views in Middle and Upper Income Groups: The Middle and Upper Income groups show a mix of sentiments, with the middle group leaning slightly positive and the upper group leaning negative.

V. Finding Context: Keywords with High Sentiment Variation

The variations between keyword sentiment and document sentiment is something to look for. It won’t likely affect a tremendous number of terms. But the ones you find, provided they are clear and compelling, can significantly impact communication strategies and content creation in several ways:

Understanding Audience Perception

The variation highlights how specific terms or concepts are perceived differently from the overall content. This can help in tailoring messages that resonate with the audience's sentiments and perceptions.

Nuanced Messaging

Recognizing that certain keywords have different sentiment values compared to the overall document can guide the development of more nuanced messages. It can help in avoiding potential misunderstandings or misinterpretations by the audience.

Content Tone Alignment

In content creation, aligning the tone of individual elements (like keywords) with the overall tone of the content is crucial for consistency. Large variations might indicate a mismatch that could confuse or mislead the audience.

Targeted Communications

By understanding which keywords diverge significantly in sentiment from the overall content, communicators can more effectively target their messages. They can emphasize or de-emphasize certain aspects to align with their communication goals.

Crisis Communication and Reputation Management

In situations where public sentiment is critical, knowing the sentiment variation can help in addressing specific concerns more effectively. For instance, a keyword with a negative sentiment in a generally positive document might need special attention in communications.

SEO and Online Visibility

For digital content, understanding keyword sentiment can influence SEO strategies. Keywords with strong sentiments (positive or negative) can affect click-through rates, engagement, and overall visibility.

Audience Engagement

Content that aligns with the audience's emotional and sentimental leanings can lead to higher engagement. Identifying sentiment variations helps in creating content that resonates more deeply with the audience.

Brand Perception

Consistency in sentiment between keywords and overall content can influence brand perception. Discrepancies might suggest a lack of coherence in brand messaging, affecting brand image.

Market Research

Variations in sentiment can offer insights into market trends and consumer attitudes, guiding content creators in developing material that is timely and relevant to their audience.

Ad Campaigns and Marketing Strategies

In advertising and marketing, leveraging the sentiment of keywords can enhance the effectiveness of campaigns. Understanding sentiment variations can guide the tone and messaging of campaigns for better audience alignment.

Categories & Concepts Analysis

Categorization in NLU is widely used for organizing and filtering large volumes of text data, such as in content management systems and in business intelligence for analyzing customer feedback and market trends. It helps in quickly identifying the relevant texts for further analysis or action.

Concepts enable us to understand the context or the broader topic of the content. They are used to categorize and theme content. Concepts represent abstract ideas that are often broader than specific instances or entities and help in grasping the meaning of text beyond keywords. 

I. Category Data Analysis

General Overview

• Total Entries: 584.

• Unique Names: 112.

• Most Frequent Name: Alexander Miller (20 occurrences).

• Household Income Levels: 5 unique levels, with "Working Class" being the most frequent (207 times).

• Story Titles: 81 unique titles, "Create a living wage" being the most common (66 times).

• Story Texts: 97 unique texts, "Childcare assistance" being the most repeated (24 times).

• Unique Categories: 106.

Category Relevance Score

• Mean: 0.69.

• Standard Deviation: 0.146.

• Range: 0.323 to 0.999.

Top Categories (by mean relevance score)

Categories such as "/society/work/unemployment" have higher occurrences and relevance scores.

II. Concept Data Analysis

General Overview

• Total Entries: 5600

• Unique Concepts: 2058

• Most Frequent Concept: "United States" (52 occurrences)

• Most Common Story Text: "Paid family leave" (250 times)

Concept Relevance Score

• Mean: 0.317

• Standard Deviation: 0.235

• Range: 0.027 to 0.995

III. Top Concepts (by mean relevance score)

Concepts like "Immigration reform" and "Tax credit" have the highest relevance scores but appear less frequently.

Insights and Trends

Categories

• The data indicates a strong focus on societal and welfare-related issues, particularly concerning the working class. This suggests a narrative centered around socio-economic challenges.

• The high occurrence of categories like "/society/work/unemployment" underscores the emphasis on employment issues in these stories.

Concepts

• The presence of diverse concepts, including specific policies like "Tax credit" and broad topics like "United States," highlights the varied thematic content within the stories.

• High relevance scores for concepts such as "Immigration reform" indicate specific stories or segments focusing intensely on these issues.

IV. Cross-Reference Analysis Report: Categories and Concepts in Economic Stories

Overview

The cross-reference analysis between categories and concepts in the economic stories dataset reveals how specific concepts are associated with broader categories. This analysis helps in understanding the thematic interplay and framing of issues within the stories.

Top Category-Concept Combinations

• The category "/society/work/unemployment" shows a strong correlation with various concepts, indicating a focused narrative around employment issues.

• Concepts such as "Contract," "Foreclosure," "English Poor Laws," "Great Recession," and "Macroeconomics" are highly relevant within this category.

Category and Concept Relevance Scores

• All top combinations have a category relevance score of 1.0, demonstrating their strong association with the stories in which they appear.

• Concept relevance scores vary, with "Contract" and "Foreclosure" having the highest scores (around 0.73), indicating their significant presence within the unemployment narratives.

Insights and Implications

• Thematic Depth: The association of concepts like "Great Recession" and "Macroeconomics" with unemployment stories suggests a depth in addressing economic issues, possibly including historical contexts and broader economic theories.

• Narrative Focus: The high relevance of legal and economic concepts within the unemployment category points to a narrative focus on the legal and economic dimensions of employment issues.

Further Interpretation

• Audience Engagement: Understanding these thematic correlations can aid in tailoring content to resonate more effectively with specific audience segments, particularly those interested in or affected by unemployment and economic policies

• Strategic Communication: For digital advertising and communication strategies, highlighting these key concepts within relevant categories could enhance audience engagement and message relevance.

This cross-reference analysis provides a nuanced understanding of the thematic structure within the economic stories, highlighting the intricate relationships between broader categories and specific concepts. This insight can be leveraged for more effective storytelling and targeted communication strategies.

V. Merging the Datasets: Story Impact Assessment Report

Overview

The story impact assessment focuses on understanding how different themes resonate with various audience segments. This is done by analyzing the average relevance scores of categories and concepts for each story.

Key Findings

Top Stories by Thematic Impact:

1. "Living wage not enough": Highest category relevance score (0.9996) and a significant concept relevance score (0.4961).

2. "Universal health car": Notably high category (0.9980) and concept (0.5374) relevance scores.

3. Other stories like "The price of groceri," "Rent control because," and "Tax-exempt status of" also show high thematic impact.

Relevance Scores

• The stories listed demonstrate a strong thematic presence in their respective categories, as indicated by the high category relevance scores.

• The concept relevance scores vary, suggesting different levels of specificity or focus on particular concepts within each story.

• Insights and Implications

Audience Engagement

• Stories like "Living wage not enough" and "Universal health car" likely resonate strongly with audiences concerned about economic policies, healthcare, and living wages.

• The varied concept relevance scores indicate different depths of exploration or emphasis on specific issues within the stories.

Strategic Communication

• For digital communication and advertising strategies, these stories can be pivotal in engaging audiences interested in socio-economic issues.

• Tailoring content and advertising to highlight these stories and their themes can enhance relevance and engagement with targeted audience segments.

Policy and Social Implications

• These stories reflect key societal concerns such as living wages, healthcare, and rent control, suggesting potential areas for policy focus or social campaigns.

Further Interpretation

• Targeted Storytelling: Leveraging these insights for targeted storytelling can create more impactful narratives that resonate with specific audience demographics.

• Digital Advertising: For platforms like Google, Facebook, and Twitter, these insights can guide ad targeting and content creation to connect with audiences based on their interests and concerns.

This story impact assessment provides a comprehensive understanding of how different themes within the economic stories resonate with various audience segments, offering valuable insights for targeted communication strategies and policy considerations.

Keyword Analysis

Keywords are specific words or phrases that hold significant meaning within a text. In NLU, these are the terms that are identified as crucial for understanding the content and context of a conversation or document.

Overview of Data

• Total Keywords Analyzed: 471

• Average Sentiment Score: -0.117 (ranging from -0.989 to 0.934)

• Average Relevance Score: 0.656 (on a scale of 0 to 1)

Sentiment Distribution

• Negative: 197 keywords

• Neutral: 140 keywords

• Positive: 134 keywords

Average Sentiment Scores: -0.740 (negative), 0.000 (neutral), 0.677 (positive)

Emotional Context

• Average Anger Score: 0.089

• Average Disgust Score: 0.065

• Average Fear Score: 0.103

• Average Joy Score: 0.257

• Average Sadness Score: 0.292

Keywords by Household Income Level

• Middle Income: Themes like childcare assistance, unemployment insurance.

• Poor: Focus on basic income, wage issues.

• Upper Income: Childcare assistance, family leave, reparations.

• Working Class: Rent control, housing, concerns about ordinary people.

• Working Poor: Basic Income, unemployment insurance, health insurance.

Implications for Public Policy and Digital Communication

• The prevalence of negative sentiments (197 keywords) suggests economic narratives often center on challenges and grievances.

• Positive sentiments (134 keywords) in areas like basic income indicate areas of hope or satisfaction.

• Emotional scores show a significant presence of sadness and joy, reflecting a range of emotional responses to economic issues.

• Keywords associated with different income levels can guide targeted policy initiatives and communication strategies. For instance, rent control and housing are more prominent among the working class, while family leave is a key issue for upper-income households.

• Digital communication strategies should be tailored to address these varied concerns, using the emotional context to resonate with the target audience.

I. Top 10 Keywords, by Emotion

II. Keyword and Entity Relationship Analysis

A Keyword-Entity Relationship Analysis involves examining how specific keywords are associated with certain entities, such as organizations, locations, or individuals. This type of analysis helps to understand the context in which keywords are used and their connection to real-world entities.

Unemployment Insurance

• Emotions: Likely to evoke sentiments like concern, anxiety, or relief.

• Entities: Labor Departments, Government Agencies

• Interpretation: The emotional response could reflect public sentiment towards government efficacy in managing unemployment benefits.

Housing

• Emotions: May involve a mix of stress, frustration, or security.

• Entities: Real Estate Agencies, Housing Authorities, Municipal Governments

• Interpretation: Emotional responses could be indicative of public satisfaction or dissatisfaction with housing policies and market conditions.

Wage

• Emotions: Possibly associated with satisfaction, fairness, or discontent.

• Entities: Employers, Labor Unions, Business Associations

• Interpretation: The sentiments here might reflect the perceived fairness of wages and the effectiveness of labor negotiations.

Family Leave

• Emotions: Likely to evoke feelings of support, stress, or concern.

• Entities: Employers, Human Resources Departments, Government Agencies

• Interpretation: Emotional tones could indicate how supportive or inadequate family leave policies are perceived.

Government

• Emotions: Can range from trust and approval to distrust and criticism.

• Entities: Federal Government, State Governments, Policy Makers

• Interpretation: The sentiment here is likely reflective of public confidence in government actions and policy decisions.

Basic Income

• Emotions: Could elicit hope, skepticism, or optimism.

• Entities: Government Agencies, Social Welfare Organizations, Policy Makers

• Interpretation: Sentiments may reflect public attitudes towards the feasibility and desirability of basic income proposals.

Childcare Assistance

• Emotions: Likely associated with relief, gratitude, or concern.

• Entities: Childcare Providers, Social Services, Government Programs

• Interpretation: The emotional aspect here might indicate the perceived adequacy and impact of childcare assistance programs.

These emotional and sentimental insights provide a deeper understanding of public perceptions and attitudes towards these economic keywords and their related entities. Such understanding is crucial for policymakers, communicators, and organizations in addressing public concerns, shaping effective messages, and designing policies that resonate emotionally with the intended audience.

III. Income-level Stories

To analyze the Economic Stories Keyword Report by household income level, I'll categorize the data based on the income levels provided in the stories and then summarize the findings. The analysis reveals distinct patterns in how different income groups discuss and emotionally respond to economic issues.

Middle Income

• Keywords and Sentiments: Focus on support measures like "childcare assistance" and "unemployment insurance". Generally, neutral sentiments.

• Emotional Insights: Moderate joy and sadness, lower anger and disgust.

Working Poor

• Keywords and Sentiments: Emphasis on fundamental issues like "Basic Income". Positive sentiments in discussing solutions.

• Emotional Insights: Higher fear and joy, reflecting concerns and hopefulness; presence of anger and sadness.

Working Class

• Keywords and Sentiments: Topics like rent control and landlord issues. Predominantly negative sentiments.

• Emotional Insights: Elevated anger and disgust, lower joy.

Unspecified Income Levels

• Keywords and Sentiments: Broad range of economic topics with neutral sentiments.

• Emotional Insights: Balanced emotional scores, indicating a general discussion.

Income Level N/A

• Keywords and Sentiments: In these stories, the focus may vary widely since the income level is not a defining factor. The sentiment can range from positive to negative, depending on the specific content of the story.

• Emotional Insights: The emotional spectrum can be diverse. Stories with N/A income levels might exhibit a wide array of emotions depending on the narrative, context, and individual perspectives of the narrators.

The inclusion of stories with N/A income levels adds another layer to our understanding. It highlights the diversity in economic narratives when income level is not a primary factor. This comprehensive analysis underscores the varied economic sentiments and emotional perspectives across different income groups, including those where income level is unspecified or not applicable. This segmentation is crucial for understanding the multifaceted nature of economic discussions and sentiments in society.

IV. Keyword Intent Analysis

The goal of a Keyword Intent Analysis is to understand the underlying intentions or purposes behind the usage of specific keywords in various contexts. This type of analysis can reveal whether keywords are used to seek information, express opinions, advocate for policy changes, or for other purposes.

Unemployment Insurance

• Initial Intent: Seeking information, expressing concerns.

• Enhanced Insights: Entity relations with government agencies and emotional tones of concern or anxiety might indicate the keyword is often used in the context of policy discussion or personal impact narratives.

Housing

• Initial Intent: Discussing affordability, advocating for policies.

• Enhanced Insights: The association with entities like housing authorities and emotions like stress or security suggests discussions around housing are both personal (impact on individuals) and policy-oriented (advocacy for change).

Wage

• Initial Intent: Debating fairness, advocating for increases.

• Enhanced Insights: The connection to employers and labor unions, combined with sentiments of satisfaction or discontent, indicates a focus on negotiation outcomes and labor rights discussions.

Family Leave

• Initial Intent: Advocacy, sharing experiences.

• Enhanced Insights: Relations with HR departments and government, along with emotions of support or stress, point towards a mix of personal experience sharing and advocacy for policy reform.

Government

• Initial Intent: Critiquing or supporting policies.

• Enhanced Insights: The link to policy makers and varied sentiments suggests usage for both support and critique of government actions, reflecting diverse political ideologies and concerns.

Basic Income

• Initial Intent: Advocacy, discussing impacts.

• Enhanced Insights: Associations with social welfare organizations and mixed emotions of hope or skepticism imply discussions around both the theoretical aspects of basic income and its practical implications.

Childcare Assistance

• Initial Intent: Seeking information, discussing challenges.

• Enhanced Insights: The correlation with childcare providers and government programs, and emotions of relief or concern, indicates discussions around the practical aspects of obtaining assistance and the systemic challenges in childcare provision.

By incorporating these additional insights, the Keyword Intent Analysis becomes more comprehensive. It not only considers the surface-level intent behind the keyword usage but also the broader context, emotional backdrop, and the entities involved. This approach provides a deeper understanding of the narratives in economic stories, useful for tailoring policy-making, research, and communication strategies to address the specific concerns and interests of different audiences.

V. Keyword Semantic Clustering

Each of the clusters below represents a distinct thematic area within the broader economic narratives present in the dataset. Understanding these clusters helps in identifying the major themes in public discourse around economic issues, providing valuable insights for policymakers, researchers, and communicators in addressing these topics effectively.

Employment and Benefits

• Keywords: Unemployment Insurance, Wage, Living Wage, Family Leave

• Analysis: This cluster revolves around employment-related benefits and earnings. It encompasses topics like unemployment support, wage levels, and family leave policies. These keywords indicate discussions focused on labor rights, compensation, and support for workers.

Housing and Real Estate

• Keywords: Housing, Rent Control

• Analysis: This cluster deals with housing affordability and regulations. It includes discussions on the general state of housing and specific policies like rent control. These keywords suggest a focus on the challenges faced by individuals in securing affordable housing and the policies that impact this sector.

Government and Policy

• Keywords: Government, Basic Income, Debt

• Analysis: This cluster captures topics related to government actions and economic policies. Keywords like basic income and debt are indicative of broader economic policy discussions, reflecting public sentiment and debates on government strategies in economic management.

Social Welfare

• Keywords: Childcare Assistance

• Analysis: This cluster is centered around social support systems, specifically childcare assistance. The presence of this keyword highlights the importance of childcare in social welfare discussions and the challenges faced by families in accessing adequate childcare support.

Entities & Relations Analysis

Entities are distinct elements within a text that represent real-world objects, concepts, or individuals. These entities are typically nouns or noun phrases and are integral for extracting meaning and context from language data.

Relations data refers to the identification and analysis of the relationships between entities within a text. This aspect of NLU involves understanding how different entities (like people, organizations, locations, etc.) are connected to each other based on the context provided in the text.

I. Entity Data Overview

This report provides a comprehensive overview of the entities present in the economic stories, reflecting a diverse range of types, from locations and organizations to emotional contexts. The dominance of emotions like sadness and joy suggests that these economic narratives evoke strong emotional responses, highlighting the personal impact of economic issues. Such insights can be valuable for understanding public sentiment and tailoring communication strategies accordingly.

Total Number of Entities: The dataset contains 80 entity instances.

Unique Entities: There are 37 distinct entities identified across the stories.

Unique Stories with Entities: Entities are distributed across 15 different stories.

Entity Type Distribution

• Location: 17 instances

• Duration: 14 instances

• Organization: 13 instances

• Money: 11 instances

• Number: 8 instances

• Date: 6 instances

• Person: 5 instances

• Percent: 4 instances

• Job Title: 2 instances

Average Entity Confidence Score: The average confidence score for entity identification is approximately 0.68. This suggests a moderate to high level of accuracy in identifying entities.

Dominant Emotion Distribution

• Sadness: 36 instances

• Joy: 30 instances

• Disgust: 12 instances

• Fear: 2 instances

II. Relations Data Overview

The "Relations" data from the "Economic Stories Entities and Relations Report" offers insights into how entities are interconnected within the context of economic stories. Here's a detailed NLU report based on the relations data:

Total Number of Relations: The dataset contains 80 instances of relations.

Unique Relation Types: There are 11 distinct types of relations identified.

Average Relation Confidence Score: The average confidence score for relation identification is approximately 0.59, indicating a moderate level of certainty in the identification of relations.

Relation Type Distribution

• Employed By: 17 instances

• Located At: 14 instances

• Resides In: 13 instances

• Part Of Many: 11 instances

• Has Attribute: 9 instances

• Parent Of: 4 instances

• Affected By: 3 instances

• Agent Of: 3 instances

• Manager Of: 3 instances

• Awarded To: 2 instances

• Member Of: 1 instance

Relation Story Context

The relations appear in various stories, with some being more prevalent in specific narratives. For example:

• 'Employed By' is most common in stories like "Make sure everyone w" (9 instances) and "The price of groceri" (8 instances).

• 'Located At' frequently appears in the story "Rent control because" (10 instances).

• 'Resides In' is prominent in "OTHER: You left out" (6 instances) and "Rent control because" (5 instances).

• Other relations like 'Part Of Many' and 'Has Attribute' are distributed across multiple stories.

This report indicates a diverse range of relation types, each contributing to the narrative of economic stories in different ways. The prevalence of certain relation types in specific stories suggests thematic patterns, like employment or location being central to certain economic discussions. Understanding these relations can provide deeper insights into the economic issues being discussed, the perspectives of the storytellers, and the emotional and social contexts they are set in. ​​

III. Entities and Relations by Income Level

The focused analysis of the "Entities and Relations" data reveals how different types of entities are connected with various relation types. These patterns provide insights into the nature of economic discussions in the dataset, highlighting the importance of locations, organizations, and financial aspects in these narratives. Understanding these connections can be vital for grasping the underlying themes and focuses of economic stories.

Here are the top 10 most common combinations:

• Location entities related through Located At: 10 instances

• Organization entities with the relation Employed By: 6 instances

• Money entities in the context of Employed By: 5 instances

• Location entities with the relation Resides In: 5 instances

• Duration entities related through Has Attribute: 4 instances

• Money entities associated with Has Attribute: 3 instances

• Number entities related through Part Of Many: 3 instances

• Number entities in the context of Employed By: 3 instances

• Person entities with the relation Resides In: 3 instances

• Duration entities associated with Employed By: 2 instances

This analysis suggests certain trends in the dataset:

• Locations are frequently mentioned in the context of where something is located (Located At).

• Organizations and Money are often discussed in employment contexts (Employed By).

• Personal entities like Persons and Locations are associated with residency (Resides In).

• Temporal (Duration) and quantitative (Number) entities are linked with attributes and broader contexts (Has Attribute, Part Of Many).

IV. Contextual Analysis of Entities

The analysis of entities, relations, and household income levels from the "Entities and Relations" data reveals the following top 10 combinations, highlighting the interaction of these aspects within different income brackets:

1. Location entities related through Located At in Working Class contexts: 10 instances.

2. Organization entities with the relation Employed By in Middle Income contexts: 6 instances.

3. Location entities with the relation Resides In in Working Class contexts: 5 instances.

4. Duration entities related through Has Attribute in Working Class contexts: 4 instances.

5. Number entities related through Part Of Many in Working Class contexts: 3 instances.

6. Number entities in the context of Employed By in Working Class contexts: 3 instances.

7. Money entities associated with Has Attribute in Working Class contexts: 3 instances.

8. Money entities in the context of Employed By in Middle Income contexts: 3 instances.

9. Money entities related through Part Of Many in Working Class contexts: 2 instances.

10. Organization entities with the relation Manager Of in Middle Income contexts: 2 instances.

This analysis suggests certain trends:

• The Working Class income level frequently appears in the dataset, especially in the context of locations (Located At, Resides In), financial aspects (Money - Employed By, Has Attribute), and quantitative aspects (Number - Part Of Many, Employed By).

• Middle Income narratives often involve organizational structures, highlighted by relations like Employed By and Manager Of with Organization entities.

The contextual analysis of entities across different economic stories from the "Entities and Relations" data reveals how certain entities are discussed in various narrative contexts. Here are the top 10 results:

1. The entity "10 months" appears 5 times in the story titled "Living wage - I am a."

2. The entity "5 years" is mentioned 5 times in "Living wage - I am a."

3. "$50,000" is referenced 4 times in the story "Make sure everyone w."

4. "Social Security" occurs 4 times in "Make sure everyone w."

5. The entity "Medicaid" appears 4 times in "Make sure everyone w."

6. The entity "June" is mentioned 4 times in the story "Paid Family Leave fo."

7. "Detroit" is discussed 3 times in "Rent control because."

8. The entity "17-22%" appears 3 times in "Rent control because."

9. "Michigan" is mentioned 3 times in the story "Rent control because."

10. The entity "ALICE 2021" occurs 3 times in "Rent control because."

These results suggest certain patterns:

• Entities like "10 months" and "5 years" are repeatedly mentioned in a story about living wages, possibly indicating a focus on time-related aspects in this economic context.

• Financial figures ("$50,000") and social programs ("Social Security," "Medicaid") are central in the narrative of "Make sure everyone w," implying a focus on income levels and social support systems.

• Geographic locations ("Detroit," "Michigan") and specific percentages ("17-22%") are key in discussions about rent control, highlighting regional focus and quantified impacts in these economic stories.

This analysis underscores the varied contexts in which specific entities appear across different economic stories, providing insights into the thematic focuses and concerns within these narratives. ​​

Here are the next ten results from the contextual analysis of entities in different economic stories:

1. "East side" appears 3 times in the story "Rent control because."

2. "Michigan" is mentioned 2 times in "Basic Income. Imagi."

3. The entity "a hundred a month" occurs 2 times in "The price of groceri."

4. "United States" is referenced 2 times in "Basic Income. Imagi."

5. The entity "UP" appears 2 times in "The price of groceri."

6. "week" is mentioned 2 times in "The price of groceri."

7. "24,000 to 75,000" is referenced 2 times in "The price of groceri."

8. The entity "150" occurs 2 times in "The price of groceri."

9. "800" is mentioned 2 times in "The price of groceri."

10. "35,000 to 40,000$" is referenced 2 times in "The price of groceri."

These results suggest:

• Geographic references like "East side" and "Michigan" are focal points in specific stories, indicating regional relevance in these economic discussions.

• Financial figures (e.g., "a hundred a month," "24,000 to 75,000," "800," "35,000 to 40,000$") are central in the narrative of "The price of groceri," possibly highlighting concerns about cost of living or income levels.

• Entities like "United States" and "UP" suggest broader geographical or national contexts in the stories, hinting at wide-reaching economic impacts or policies.

The repeated mention of these entities in specific stories points to one of two things. One, the thematic emphasis and key concerns within these narratives offer deeper insights into the economic issues being discussed. Two, there may not be enough data to properly support results outside of the top ten. Remember, when using this data stick to the clearest, most compelling connections possible.

Subject/Verb/Object Analysis

Subject/verb/object analysis is a fundamental part of NLU, serving as a building block for more complex language understanding tasks. It provides a basic framework for parsing sentences and extracting meaning.

I. Overview

Frequency and Distribution

• Subjects: 'Government', 'living wage', and 'people' are the most frequent subjects.

• Verbs: The verb 'be' is used most often, followed by 'regulate' and 'help'.

• Objects: 'Amount' and 'housing' are common objects, along with 'utility companies'.

Relationships and Patterns

• Subject-Verb Pairs: The pair 'government-control' and 'government-regulate' are prevalent.

• Verb-Object Pairs: 'Regulate-utility companies' and 'provide-safety net' are notable combinations.

• Subject-Object Pairs: The combination of 'government-housing' and 'government-amount' are frequently observed.

Contextual Analysis

• Unique SVO combinations like 'government-control-housing' and 'government-regulate-amount' indicate a focus on government's role in regulating housing and financial matters.

Emotion and Sentiment Correlation

• Emotion-Subject Correlation: 'Joy-living wage' and 'disgust-government' are prominent, suggesting positive sentiment towards living wage and negative towards government actions.

• Emotion-Verb Correlation: 'Joy-be' and 'sadness-be' are common, reflecting mixed emotions in the context of being or existence.

• Emotion-Object Correlation: 'Sadness-utility companies' and 'disgust-amount' indicate negative sentiments associated with financial burdens.

Here is a list of the top unique subjects mentioned in the economic stories, along with their frequency of occurrence:

Government - 12 mentions

Living Wage - 11 mentions

People - 10 mentions

Wages - 6 mentions

Costs - 5 mentions

Basic Income - 4 mentions

Annual Income - 4 mentions

Higher Wages - 4 mentions

Wage - 3 mentions

Time - 3 mentions

These subjects represent the most frequently discussed topics in the economic narratives, indicating their significance in the stories.

This analysis suggests a strong focus on governmental roles, financial challenges, and mixed emotions towards economic conditions in the stories. The frequent appearance of terms like 'living wage', 'housing', and 'utility companies' underlines economic concerns prevalent in the narratives. The emotions associated with these terms also reflect public sentiment towards these issues.

II. Frequency Distribution

III. Relationships & Patterns

A deeper analysis into Relationships and Patterns in the Subject-Verb-Object (SVO) data provides the following insights. There are not many of these, so make sure that the connections are clear and compelling before using.

Common SVO Sequences

• Government-Control-Housing (3 occurrences)

• Government-Regulate-Housing (3 occurrences)

• Government-Regulate-Amount (3 occurrences)

• Government-Control-Amount (3 occurrences)

• New Child-Take-Care of a Family Member (1 occurrence)

• Main Goal-Be-Medicaid (1 occurrence)

• Main Goal-Be-People (1 occurrence)

• Main Source of Support-Be-Social Security Disability (1 occurrence)

• Minimum Wage Jobs-Increase-Wages (1 occurrence)

• Money-Live-Love (1 occurrence)

These combinations reflect the emphasis on government actions in housing and financial regulation, and highlight specific scenarios like the challenges faced by new parents and those relying on social security disability. 

Impactful SVO Combinations (High Emotion Score)

• Basic Income-Create-Living Wage Rent Control

• Lower Fees-Regulate-Utility Companies

• Fines-Regulate-Utility Companies

• Housing-Be-Michigan

• Housing-Be-Issue

• Income-Be-Stretch

• Internet-Be-Month

• Low Delayed Tax Payments-Regulate-Utility Companies

• Low Income Families-Regulate-Utility Companies

• Money-Live-Love

These combinations, associated with higher emotional scores, indicate significant public sentiment. They involve topics like basic income, utility regulation, and the stretching of income, which are critical in public policy considerations.

Specific Subject-Verb-Object Patterns

• This analysis mirrors the common SVO sequences, emphasizing government's role in controlling and regulating housing and financial matters. It also highlights the nuanced challenges in areas like minimum wage jobs, family care, and social support systems.

Actionable Insights

• The government's role in housing and financial regulation is a major focus, suggesting the importance of these areas in public policy and communication strategies.

• Emotional responses are strongly tied to economic issues like basic income, utility regulation, and income management, indicating areas of high public interest or concern.

• Understanding these patterns can aid in developing targeted policies and communication campaigns that address these key issues, resonating with public sentiment and concerns.

IV. Contextual Analysis

V. Emotion and Sentiment Correlation

The Emotion and Sentiment Correlation analysis reveals how different emotions are associated with specific Subject-Verb-Object (SVO) combinations in the economic stories.

Emotion Correlation with Subjects

• Living Wage - Joy: Mentioned 9 times, indicating positive sentiment towards living wage initiatives.

• Government - Disgust: Appeared 8 times, suggesting a negative sentiment towards government actions or policies.

• People - Joy: 5 mentions, reflecting positive sentiment in stories involving people.

• Costs - Sadness: 5 mentions, indicating concerns or negative feelings about rising costs.

• Wages, Annual Income, Higher Wages - Joy: Each 4 mentions, showing a positive reaction to topics related to wages and income.

• Government - Joy: 4 mentions, indicating some positive sentiment towards government in certain contexts.

• Time, Wage - Joy: 3 mentions each, suggesting a positive outlook on these economic aspects.

Emotion Correlation with Verbs

• Be - Joy: 18 occurrences, showing a general positive sentiment in narratives.

• Be - Sadness: 11 occurrences, reflecting a more negative or challenging aspect in certain narratives.

• Help - Joy: 10 mentions, indicative of positive sentiment towards supportive actions or policies.

• Regulate, Make - Sadness: 5 mentions each, possibly reflecting challenges or issues with regulation and making ends meet.

• Provide - Joy: 5 mentions, showing positivity towards provision or support.

• Allow - Joy, Control/Regulate - Disgust: 4 mentions each, indicating varied sentiments towards control and regulation.

• Make - Joy: 3 mentions, reflecting positive outcomes or situations.

Emotion Correlation with Objects

• Utility Companies - Sadness: 5 mentions, indicating concerns or negative feelings towards utility companies.

• Issue - Sadness, Amount/Housing - Disgust: 4 mentions each, reflecting negative sentiments towards these financial and housing issues.

• Safety Net - Joy: 4 mentions, showing a positive reaction to safety nets.

• Lot - Joy, Care of a Family Member - Sadness: 3 mentions each, indicating varied sentiments towards family care and other issues.

• Control - Anger, Shelter - Joy: 2 mentions each, showing strong emotions related to control and positive sentiment towards shelter.

Actionable Insights

• Public Sentiment: There is a clear emotional response to economic terms, particularly around living wages, government actions, and utility companies, which can guide policy and communication strategies.

• Policy Implications: The mixture of joy and disgust towards government and sadness related to costs and utility companies highlights areas for policy focus and improvement.

• Communication Strategies: Understanding these emotional correlations can help in crafting messages that resonate with the public's feelings towards these economic issues.

VI. Subjects

The Natural Language Understanding (NLU) analysis of the subject data from the economic stories provides four key insights:

1. Subject Frequency and Prominence

• Top Subjects: 'Government' (12 mentions), 'Living Wage' (11), and 'People' (10) are the most frequently mentioned subjects. This suggests these are key areas of focus in the economic narratives.

• Other notable subjects include 'Wages', 'Costs', 'Basic Income', 'Annual Income', and 'Higher Wages'.

2. Subject Context

• Basic Income: Associated with stories titled "Basic Income Create", showing a focus on the creation or implementation of basic income policies. The dominant emotion is 'Joy'.

• Black America: Appears in stories related to "Create a living wage", indicating a connection between living wage discussions and Black communities. The dominant emotion is 'Joy'.

• Cost of Living: Found in narratives about the challenges of living costs, with a dominant emotion of 'Sadness'.

• Higher Wages: Linked with positive narratives about increasing wages, with a dominant emotion of 'Joy'.

3. Subject-Emotion Correlation

• Living Wage - Joy: The most common correlation, suggesting positive sentiment towards living wage initiatives.

• Government - Disgust: Indicates a significant level of discontent or criticism towards governmental actions or policies.

• Costs - Sadness: Reflects concerns or challenges related to rising living costs.

4. Subject-Category Relation

• Government and Rent Control: A frequent pairing, showing a focus on government policies regarding rent control.

• Living Wage and Wage Discussions: Living wage is often discussed in the context of creating or advocating for higher wages.

• Costs and Rent Control: Indicates a connection between living costs and rent control policies.

Actionable Insights

• Policy Focus: Subjects like government, living wage, and costs are central to the economic stories, highlighting key areas for policy development and public discourse.

• Emotional Resonance: Positive sentiments towards 'Living Wage' and 'Higher Wages' contrast with negative emotions towards 'Government' actions and the 'Costs' of living, suggesting areas of public satisfaction and dissatisfaction.

• Communication Strategies: Understanding the emotional context and prevalent themes associated with these subjects can guide effective communication and policy-making strategies.

VII. Subject Contexts

To expand on the contextual associations of the top subjects from the economic stories, we'll dive deeper into the contexts in which these subjects are mentioned. This includes exploring the stories they are associated with, the narrative themes, and the emotions tied to each subject. This analysis helps in understanding the broader narrative and thematic contexts of these subjects:

• Government: Often mentioned in the context of policies and regulations. The association with both positive and negative emotions suggests a complex public sentiment towards government actions and decisions.

• Living Wage: Typically discussed in stories advocating for fair wages or highlighting income disparities. The positive emotion associated with this subject indicates general support for living wage policies.

• People: This subject appears in a variety of contexts, reflecting diverse aspects of society and economic conditions. The emotional range suggests different impacts of economic policies or situations on people.

• Wages: Frequently comes up in discussions about employment, economic stability, and labor rights. The sentiment here could reflect public attitudes towards wage levels and fairness.

• Costs: Often associated with the challenges of managing living expenses. The presence of sadness in these contexts likely points to the stress and difficulties people face with rising costs.

• Basic Income: This subject is commonly found in narratives discussing poverty alleviation and social security measures. Positive emotions in these contexts might indicate approval or optimism towards basic income initiatives.

• Annual Income: Appears in discussions about financial planning, economic status, and wealth distribution. The sentiment here can provide insights into public attitudes towards income levels and economic inequality.

• Higher Wages: Typically associated with advocacy for increased pay and improved living standards. The positive emotions linked to this subject suggest support for higher wage policies.

• Wage: This subject is often part of broader discussions on employment and remuneration. The emotions connected to 'Wage' can reveal public sentiments about pay scales and worker compensation.

• Time: This subject might appear in various contexts, possibly relating to time management, work-life balance, or economic changes over time. The emotional undertones here could reflect the value or challenges associated with time in an economic sense.

Expanded Insights

• Each subject carries specific thematic and emotional weights, reflecting their role in the economic stories.

• Understanding these contextual associations is crucial for grasping the broader narrative threads and public sentiments in these stories.

• These insights can inform targeted communication strategies, policy-making, and social initiatives that resonate with public concerns and aspirations.

VIII. Subject Entities and Relations

The Enhanced Entities and Relations report from the economic stories provides a multifaceted view of the narratives. Here's an overview of the data:

General Information

The report contains 80 entries.

It comprises 12 columns with a mix of categorical and numerical data.

The data includes information about names, income levels, story titles, texts, entities, entity types, and emotions.

Key Columns Overview

• Name: Names of individuals mentioned in the stories.

• Household Income Level: Income levels of the households mentioned, predominantly 'Working Class'.

• Story Title: Titles of the stories, with "The price of groceries" appearing frequently.

• Story Text: The actual text of the stories.

• Entities: Identified entities within the stories, such as "UBI" or "Detroit".

• Entity Type: Types of entities, like 'Organization' or 'Location'.

• Entity Confidence Score: A numerical score representing the confidence level in the entity identification.

• Relations Sentence: Sentences from the story text that illustrate the relationships between entities.

• Relation Type: Types of relationships, such as 'partOfMany' or 'locatedAt'.

• Relation Score: A score indicating the strength of the identified relation.

• Dominant Emotion: The primary emotion detected in the story segment.

• Dominant Emotion Score: A score representing the intensity of the dominant emotion.

Sample Data Insights

• The report shows a focus on organizations and locations, as indicated by entities like "All Americans Student Loan Debt" and "Detroit".

• Relation types like 'partOfMany' and 'locatedAt' suggest connections between entities and broader themes or locations.

• The dominant emotions in these stories are 'sadness' and 'joy', indicating a range of emotional responses to economic issues.

• This report offers valuable insights into the entities involved in the economic narratives, their relationships, and the emotional context of these stories. It can provide a deeper understanding of the key actors, locations, and organizations mentioned in the stories and how they relate to the overall economic themes and sentiments.

IX. Subject Thematic Correlations

This analysis will highlight the key themes associated with each subject and provide insights into their relevance in the context of the stories:

• Government: Often correlated with themes of regulation, policy-making, and governance. This includes discussions about government's role in economic matters such as rent control, welfare policies, and regulatory actions.

• Living Wage: Linked with themes of fair compensation, labor rights, and economic justice. Stories featuring this subject often delve into the impact of wages on quality of life and the push for equitable pay standards.

• People: A broad subject that correlates with a range of themes, from individual stories of hardship and success to broader societal issues like unemployment and social welfare. It reflects the human aspect of economic narratives.

• Wages: Tied to themes of employment conditions, income stability, and labor market trends. Discussions often focus on the adequacy of wages in meeting living expenses and maintaining a decent standard of living.

• Costs: Associated with themes of financial burden, cost of living, and economic pressures faced by individuals and families. This subject often highlights the challenges of managing daily expenses against stagnant incomes.

• Basic Income: Correlates with themes of social security, poverty reduction, and economic reform. Narratives with this subject tend to explore the potential of basic income as a tool for addressing economic disparities.

• Annual Income: Linked with themes of financial planning, wealth distribution, and economic stratification. Stories featuring this subject often address issues of income inequality and the varying economic experiences across different income levels.

• Higher Wages: Closely associated with labor advocacy, economic empowerment, and the fight for better pay. This subject is often at the forefront of narratives discussing wage gaps and the movement for higher minimum wages.

• Wage: Encompasses themes of compensation fairness, economic value of labor, and worker rights. It's a central subject in discussions about the relationship between work and remuneration.

• Time: A versatile subject, often correlated with themes of economic change over time, time management in professional settings, and work-life balance. It can also feature in narratives about long-term economic planning and trends.

Expanded Insights

• These thematic correlations provide a lens to understand the complexities and multifaceted nature of economic issues.

• Each subject serves as a gateway to exploring deeper economic and social themes, reflecting the interconnectedness of these topics.

• Recognizing these correlations is vital for comprehending the broader economic discourse and can aid in formulating more nuanced responses and policies that address these interlinked themes.

X. Subject Emotion Dynamics

Expanding on the Subject-Emotion Dynamics from the economic stories, we'll delve deeper into how each subject is emotionally perceived or associated within the narratives. This analysis helps to understand the emotional impact and sentiment linked to each subject, offering insights into public feelings and attitudes towards these economic topics:

• Government: Evokes mixed emotions, including both 'Disgust' and 'Joy'. This reflects a complex public sentiment towards government actions, possibly indicating approval of certain policies and disapproval of others.

• Living Wage: Strongly associated with 'Joy', indicating a generally positive public sentiment towards discussions about living wages. This suggests support for initiatives aimed at ensuring fair compensation.

• People: Linked with 'Joy', highlighting a positive emotional response to stories focused on the populace. This could reflect narratives that emphasize empowerment, resilience, or success.

• Wages: Also associated with 'Joy', suggesting a positive perception of discussions around wages, possibly in the context of wage increases or fair compensation.

• Costs: Correlated with 'Sadness', indicating the emotional challenges or stress associated with managing living costs. This reflects public concerns over financial burdens.

• Basic Income: When this subject appears, it's often linked with positive emotions like 'Joy', possibly reflecting optimism or approval of basic income as a potential solution to economic disparities.

• Annual Income: Associated with 'Joy', which might indicate positive sentiments towards discussions about income levels, possibly in the context of income growth or financial stability.

• Higher Wages: This subject's association with 'Joy' suggests support for narratives advocating higher wages, reflecting public approval of efforts to improve wage standards.

• Wage: The emotional response to 'Wage' is predominantly 'Joy', indicating a positive perception of wage-related discussions, perhaps in contexts of wage fairness or increases.

• Time: Linked with 'Joy', suggesting a positive response to narratives involving time, possibly in contexts like time management, work-life balance, or long-term economic growth.

Expanded Insights

• The emotional dynamics around these subjects provide a window into public sentiments and reactions to various economic issues.

• Understanding these dynamics is crucial for grasping the emotional underpinnings of economic narratives, which can significantly influence public opinion and policy reception.

• Recognizing the emotional context of these subjects can guide effective communication strategies and policy formulation, ensuring resonance with public sentiment and addressing underlying emotional concerns.

XI. Verbs

The Natural Language Understanding (NLU) report on verbs from the economic stories reveals four key insights into how actions and processes are represented in the narratives:

1. Frequency and Distribution of Verbs

The most frequent verbs are 'be' (32 mentions), 'regulate' (11), 'help' (10), 'make' (8), 'control' (6), 'pay' (6), 'take' (5), 'provide' (5), 'create' (5), and 'allow' (4).

This frequency indicates the prominence of certain actions and states in the economic narratives.

2. Contextual Associations

Verbs like 'meet', 'receive', 'live', 'contract', and 'have' are associated with strong emotional responses, such as 'joy' and 'sadness'.

'Meet' is closely linked with 'joy', suggesting positive outcomes or achievements.

'Receive', 'contract', and 'deny' are associated with 'sadness', indicating challenges or unmet needs.

3. Verb-Emotion Correlation

'Be' is predominantly associated with 'joy' and 'sadness', indicating its use in a variety of contexts with mixed emotional undertones.

'Help' is strongly linked with 'joy', reflecting positive actions or support.

'Make', 'regulate', and 'control' associated with 'sadness' and 'disgust', possibly indicating challenges or contentious actions.

4. Verb-Thematic Correlation

The verb 'be' is frequently used in stories about creating a living wage, reflecting a state or condition.

'Help' and 'provide' appear in narratives about living wages and basic income, suggesting supportive or enabling actions.

'Regulate', 'make', and 'control' are often found in stories about rent control and tax rates, indicating governance and management actions.

Expanded Insights

• The verbs used in the economic stories reveal the nature of actions, states, and processes discussed in the narratives.

• Understanding the frequency, context, and emotional associations of these verbs provides deeper insights into the narrative dynamics and thematic focuses of the stories.

• These insights can inform communication strategies, policy-making, and public discourse by highlighting the actions and processes that resonate most with the public and their emotional implications.

XII. Verb Tense

The detailed Natural Language Understanding (NLU) report on verb tense from the economic stories provides insights into how different verb tenses are used and perceived in the narratives:

1. Frequency and Distribution of Verb Tenses

• Present Tense: Most frequent with 69 mentions.

• Infinitive: 39 mentions.

• Past Tense: 11 mentions.

• Present Participle: 9 mentions.

• Past Participle: 3 mentions.

2. Contextual Associations

• Present Participle: Associated with a variety of stories, evoking emotions like sadness, joy, and disgust. This suggests dynamic or ongoing actions in the narratives.

• Past Tense: Used in stories that evoke sadness, joy, and anger, indicating reflections on past events or conditions.

• Present Tense: Predominantly found in narratives about current situations or states, associated with a range of emotions including joy, fear, sadness, and disgust.

• Past Participle: Appears in contexts that evoke joy and anger, possibly indicating completed actions with significant emotional impact.

• Infinitive: Used in a range of contexts, evoking mixed emotions. This tense often indicates intentions or potential actions.

3. Verb Tense-Emotion Correlation

• Present Tense: Strongly correlated with 'joy' and 'sadness', reflecting a mix of positive and negative sentiments in current situations.

• Infinitive: Associated with a variety of emotions, highlighting its use in expressing intentions or possibilities with emotional undertones.

4. Verb Tense-Thematic Correlation

• The use of different tenses is correlated with specific thematic contexts in the stories. For instance:

• Present Tense in stories about living wages and rent control.

• Past Tense in narratives reflecting on past conditions or events like cost of living.

• Infinitive in stories about future-oriented topics like basic income creation.

Expanded Insights

• The choice of verb tense in the stories reflects the temporal context of the narratives, whether they are focusing on past events, current states, or future possibilities.

• Understanding these verb tense dynamics can provide deeper insights into the storytelling approach, the temporal framing of economic issues, and the emotional impact these narratives have on the audience.

• Recognizing the correlation between verb tenses and thematic or emotional contexts can aid in crafting more effective narratives and communication strategies, aligning the temporal framing of the message with the desired emotional impact.

XII. Impact of Verb Tense on Story Perception

The Impact of Verb Tense on Story Perception is a fascinating aspect of narrative analysis, especially in economic storytelling. Verb tense plays a crucial role in shaping how readers perceive the immediacy, relevance, and relatability of a story. 

The examples below demonstrate how different verb tenses convey varying degrees of immediacy, reflectiveness, and potentiality. The present participle and present tense offer a sense of current involvement and ongoing process, while the past tense provides historical context and reflection. The past participle connects past actions with present relevance, and the infinitive opens up possibilities and intentions for the future. Each tense thus contributes uniquely to shaping the narrative's temporal setting and emotional depth.

We want to take a moment to explain how this works. Let's explore how different verb tenses can influence story perception in economic narratives:

1. Present Tense ("is", "does", "manages")

• Immediacy and Urgency: Present tense gives a sense of immediacy, making the story feel current and urgent. Economic issues discussed in the present tense can seem more pressing and relevant to the reader.

• Engagement: Using the present tense can engage readers more directly, making them feel like they are part of the ongoing story.

• Example: "The economy is struggling, leading to increased unemployment rates." This makes the situation feel active and ongoing.

2. Past Tense ("was", "did", "managed")

• Reflection and Context: Past tense is often used to provide background or context to the current economic situation. It allows readers to reflect on how past events have shaped the present.

• Narrative Distance: While it can create a sense of narrative distance, it's crucial for building the story's foundation and offering a historical perspective.

• Example: "The economy was booming five years ago, with low unemployment rates." This positions the current situation in a historical context.

3. Future Tense ("will be", "will do", "will manage")

• Expectations and Predictions: Future tense is used for making projections or expressing expectations about the economic future. It introduces elements of speculation and anticipation.

• Planning and Possibility: It can instill a sense of planning and possibility, making the narrative forward-looking and focused on potential outcomes.

• Example: "The economy will recover, with predictions of job growth." This offers a forward-looking perspective, focusing on potential developments.

4. Perfect Tenses (Present Perfect, Past Perfect)

• Completed Actions or Ongoing Effects: Perfect tenses are often used to indicate actions completed in the past but relevant to the present or to show the progression of events.

• Layered Understanding: They provide a layered understanding of events, showing how past actions continue to influence the present.

• Example: "The government has introduced new economic policies." This suggests recent actions with ongoing relevance.

Impact on Story Perception

• Temporal Framing: The choice of tense frames the story temporally, affecting how readers perceive the timeline and flow of events.

• Emotional Response: Different tenses can evoke different emotional responses, with present tense often creating a sense of urgency, past tense a reflective mood, and future tense a sense of anticipation or anxiety.

• Reader Engagement: Tenses can either draw readers into the immediacy of the action or provide a more detached, reflective viewpoint.

In summary, the verb tense used in economic storytelling significantly impacts how readers perceive and engage with the narrative. It influences the sense of time, urgency, and emotional resonance, making it a critical element in effective economic narrative construction.

Integrating examples from the data will provide a clearer illustration of how verb tense impacts story perception. Let's use specific text examples from the economic stories to demonstrate the effect of different verb tenses:

Present Participle ("-ing" form)

Example: "o Paid Family Leave: policies that enable workers caring for ailing family members..."

Analysis: This present participle form suggests ongoing action and current relevance. It adds a sense of immediacy and continuous process, reflecting the evolving nature of the policy discussion.

Past Tense

Example: "Cost of living were family and live without struggle..."

Analysis: The past tense here reflects on previous living conditions, evoking a reflective mood and possibly nostalgia or regret. It situates the narrative in a past context, inviting readers to consider how things have changed or remained the same.

Present Tense

Example: "Living wage, since if wages actually kept up with inflation..."

Analysis: The use of the present tense makes the discussion about wages feel current and urgent. It directly engages the reader with the immediacy of the issue, emphasizing its ongoing relevance.

Past Participle

Example: "Living wage, since if wages actually kept up with inflation..."

Analysis: Although this is the same text as the present tense example, the past participle "kept" implies a completed action with ongoing relevance. It suggests a historical perspective that continues to impact the present.

Infinitive

Example: "Basic Income Create a living wage Rent control..."

Analysis: The use of the infinitive here indicates intentions or potential future actions. It introduces an element of planning or aspiration, focusing on what could or should be done.

XIII. Verbs & Objects

Finding clear signals supported by relevence scores and old-fashioned logic is how you decide which combination of subject, verb and object indicates intent. Then you need to find that intent across a significant enough number of verbs in order to consider it a trend. The key to this is the amount of data (in this case stories) you have to work with.

The verb/object pairs that present themselves as the most relevant in numbers become the intent you're looking for. 

XIV. Verb-Emotion Interaction Analysis

This involves examining how specific verbs contribute to the emotional tone of the narratives in the economic stories. We'll look at examples and explore how these verbs shape the emotional context.

For this analysis, we'll consider a few key verbs identified earlier as frequent in the narratives, such as "be", "regulate", "help", "make", and "control". We will analyze:

• How these verbs are used in different contexts within the stories.

• The emotional tone associated with each usage.

• Examples from the text that illustrate these dynamics.

Let's start with the verb "be" and proceed through the list. I'll extract relevant examples and insights from the SVO data.

The verb "be" is used in various contexts within the economic stories, contributing to different emotional tones. Here are a few examples illustrating this:

Context: Discussing a living wage as a basic right.

• Text: "Living wage. I feel like a living wage is a basic right..."

• Emotion: Sadness

• Emotion Score: 0.496

Analysis: Here, "be" is used to emphasize the necessity of a living wage, evoking a sentiment of sadness, possibly due to the lack of this basic right being met.

Context: Advocating for basic income in Detroit.

• Text: "Basic income. I want basic income in Detroit because..."

• Emotion: Joy

• Emotion Score: 0.426

Analysis: In this instance, "be" contributes to a hopeful or positive outlook towards the prospect of basic income, reflecting joy or optimism.

Context: Highlighting the importance of a living wage.

• Text: "Create a living wage. its important because a living wage..."

• Emotion: Joy

• Emotion Score: 0.458

Analysis: The use of "be" in this context underlines the importance of a living wage, aligning with a sense of joy or approval of this idea.

These examples demonstrate how the verb "be" can be central in shaping the emotional narrative of economic stories. Its use ranges from expressing essential needs and rights (often with a tone of sadness or urgency) to highlighting positive aspirations or approvals (associated with joy or optimism). This versatility in emotional tone underscores the verb's pivotal role in these narratives.

The verb "be" is used in various contexts within the economic stories, contributing to different emotional tones. Here are a few examples illustrating this:

Context: Discussing a living wage as a basic right.

• Text: "Living wage. I feel like a living wage is a basic right..."

• Emotion: Sadness

• Emotion Score: 0.496

Analysis: Here, "be" is used to emphasize the necessity of a living wage, evoking a sentiment of sadness, possibly due to the lack of this basic right being met.

Context: Advocating for basic income in Detroit.

• Text: "Basic income. I want basic income in Detroit because..."

• Emotion: Joy

• Emotion Score: 0.426

Analysis: In this instance, "be" contributes to a hopeful or positive outlook towards the prospect of basic income, reflecting joy or optimism.

Context: Highlighting the importance of a living wage.

• Text: "Create a living wage. its important because a living wage..."

• Emotion: Joy

• Emotion Score: 0.458

Analysis: The use of "be" in this context underlines the importance of a living wage, aligning with a sense of joy or approval of this idea.

These examples demonstrate how the verb "be" can be central in shaping the emotional narrative of economic stories. Its use ranges from expressing essential needs and rights (often with a tone of sadness or urgency) to highlighting positive aspirations or approvals (associated with joy or optimism). This versatility in emotional tone underscores the verb's pivotal role in these narratives.

Let's look at a few other interactions.

XV. Comparative Analysis of Verb Usage

For a Comparative Analysis of Verb Usage, we'll examine how similar verbs are used in different contexts within the economic stories, and explore the nuances in narrative style or thematic emphasis that arise from their usage. This analysis will help us understand the subtle differences in how verbs convey meanings, emotions, and intentions in various narrative settings.

We'll focus on a set of verbs that have similar meanings or functions but are used in different contexts. For this analysis, let's select the verbs "help" and "support", as well as "make" and "create". These pairs have related meanings but can be used distinctively in different narrative contexts.

Comparative Analysis: "Help" vs. "Support"

Usage Contexts:

• "Help": Often used in contexts of providing assistance or enabling positive outcomes.

• "Support": Typically employed in scenarios of backing, upholding, or sustaining actions or policies.

Emotional Tones:

• "Help": Generally associated with positive emotions, indicating immediate or direct assistance.

• "Support": May carry a more neutral or sustained emotional tone, suggesting ongoing backing or endorsement.

Narrative Impact:

• "Help": Conveys a sense of active involvement or intervention.

• "Support": Implies a steadier, perhaps less direct form of assistance.

Comparative Analysis: "Make" vs. "Create"

Usage Contexts:

• "Make": Often indicates causing something to happen or come into existence.

• "Create": Typically used for bringing something new into existence, often with a sense of innovation or originality.

Emotional Tones:

• "Make": Can have a wide range of emotional tones, from positive to negative, depending on the context.

• "Create": Often associated with positivity, innovation, and the start of something new.

Narrative Impact:

• "Make": Implies a transformation or change, potentially forceful or decisive.

• "Create": Suggests the initiation of something new or original, often with a constructive or creative connotation.

"Regulate" vs. "Control"

Regulate:

• Often used in the context of setting standards or rules.

• Implies a more systematic, orderly approach to management.

• Could evoke emotions related to fairness, order, or bureaucracy.

Control:

• Generally implies having command or dominance over something.

• Can evoke stronger emotional responses, such as security or oppression.

• Often used in contexts implying direct, hands-on management.

"Provide" vs. "Supply"

Provide:

• Suggests making something available or furnishing needs.

• Often carries a connotation of care or support.

• Can be used in contexts highlighting welfare, assistance, or availability.

Supply:

• Typically used in the context of furnishing or equipping with necessary items.

• More transactional or logistical in tone.

• Often found in discussions about market dynamics, resource allocation, or trade.

"Increase" vs. "Grow"

Increase:

• Indicates a rise in quantity or size.

• Often used in more technical or quantitative contexts.

• Can be associated with discussions around metrics, statistics, or explicit changes.

Grow:

• Implies a more organic or gradual process of becoming larger or more extensive.

• Often used in contexts of development, expansion, or natural progression.

• Can evoke a sense of nurturing, development, or organic change.

Through this comparative analysis, we can see how similar verbs are nuanced in their usage, contributing differently to the storytelling, emotional landscape, and thematic depth of the economic narratives. This understanding can be particularly valuable in crafting effective communication, enhancing narrative styles, and understanding the subtleties of language in economic storytelling.

XVI. Verb-Driven Story Development

Focusing on Verb-Driven Story Development, we'll explore how the use of specific verbs drives the progression of economic stories, particularly at key turning points or climactic moments. Verbs are the engines of sentences, propelling narratives forward and imparting action, change, and dynamism. In economic stories, verbs not only describe actions but also convey motivations, outcomes, and shifts in situations.

Let's examine how certain verbs catalyze narrative development in economic storytelling:

1. "Impact" and "Affect"

• Role in Story: These verbs often introduce the consequences or effects of economic policies or situations.

• Narrative Development: When a story pivots around "impact" or "affect", it typically leads to a discussion of outcomes, repercussions, or the need for response, moving the story from description to analysis or action.

2. "Implement" and "Execute"

• Role in Story: Indicate the initiation of actions or plans, especially regarding policies or strategies.

• Narrative Development: The use of these verbs often marks the transition from planning or deliberation to action, signifying a move towards realization of goals or resolutions in the narrative.

3. "Increase" and "Decrease"

• Role in Story: Used to describe changes in economic metrics like prices, wages, or investments.

• Narrative Development: These verbs can signal turning points in the economy or a character's circumstances, leading to phases of growth or decline, and prompting reactions or adaptations in the story.

4. "Negotiate" and "Bargain"

• Role in Story: Often appear in contexts of conflict resolution, deal-making, or policy formulation.

• Narrative Development: Their usage usually introduces elements of dialogue, compromise, or conflict, moving the narrative into a space of interaction and resolution between different parties.

5. "Transform" and "Revolutionize"

• Role in Story: Indicate significant changes or shifts, often in the context of innovation or reform.

• Narrative Development: These verbs often mark key moments of change or turning points in the story, leading to new phases or directions in the narrative arc.

In summary, the way verbs are employed in economic stories can significantly influence their direction and development. They not only describe actions but also set the pace, introduce new phases, and signal shifts in the narrative, making them crucial tools in effective storytelling. Understanding the role of verbs in story development allows for a deeper appreciation of the narrative craft in economic contexts.

XVII. Objects

Based on the emotional analysis of objects from the SVO dataset, here is a report highlighting the top objects with their associated dominant emotions and average emotion scores:

Object: "their basic needs"

• Dominant Emotion: Joy

• Average Emotion Score: 0.9318

Object: "a new child"

• Dominant Emotion: Sadness

• Average Emotion Score: 0.8336

Object: "compensation"

• Dominant Emotion: Sadness

• Average Emotion Score: 0.8336

Below, we look closer at these objects:

Their Basic Needs" - Joy (Emotion Score: 0.931832)

This object, associated with the highest joy score, suggests a positive sentiment towards fulfilling basic needs. This could indicate satisfaction or optimism in contexts where basic needs are being met or discussed.

"A New Child" - Sadness (Emotion Score: 0.833634)

The association of sadness with "a new child" might reflect concerns about challenges faced by families, such as financial strain, childcare issues, or societal support for new parents.

"Compensation" - Sadness (Emotion Score: 0.833634)

Similar to "a new child", the object "compensation" is also associated with sadness. This could highlight public concerns about inadequate compensation, wage issues, or the financial difficulties of workers.

"Care of a Family Member Who is Ill" - Sadness (Emotion Score: 0.833634)

This object's association with sadness likely points to the emotional and financial burdens of caring for ill family members, possibly indicating gaps in healthcare support or the impact of such responsibilities on families.

"Peace and Love" - Joy (Emotion Score: 0.819453)

The high joy score for "peace and love" indicates positive sentiment, possibly in contexts discussing societal harmony, community support, or successful social policies.

XVIII. Variety in Object Emotional Response

To analyze the variety in emotional responses we look at how different objects in the SVO dataset evoke diverse emotions:

Complexity of Emotional Responses

• The dataset shows that economic subjects are not simply associated with a single type of emotion but can evoke a range of responses. For example, while "peace and love" evoke joy, "compensation" and "family care" evoke sadness. This diversity in emotional responses highlights the multifaceted nature of how economic issues are perceived and felt by the public.

Contrasting Emotions in Similar Contexts

• Even within similar economic contexts, contrasting emotions can be observed. For instance, "their basic needs" is linked with joy, possibly reflecting satisfaction or relief in contexts where these needs are being addressed. 

• On the other hand, "a new child" and "care of a family member who is ill", which also relate to basic human experiences, are associated with sadness, indicating stress or concern about support systems and resources.

Emotion as a Reflection of Underlying Concerns

• The emotional responses are not just reactions to the objects themselves but are reflective of deeper societal and economic concerns. The sadness associated with "compensation" and "a new child" might point to worries about financial stability, social support, or the challenges of balancing work and family life.

Positive Emotions and Policy Implications

• Objects associated with positive emotions like joy, such as "peace and love", suggest areas where public sentiment is favorable. This could indicate satisfaction with certain policies or aspects of society, providing valuable feedback for policymakers and community leaders.

Emotion Scores as Indicators of Intensity

• The emotion scores associated with each object also provide an indication of the intensity of these emotions. Higher scores suggest stronger emotional responses, which can be critical in prioritizing areas for policy intervention or community support.

This analysis of the variety in emotional responses underscores the importance of understanding the emotional dimension of economic discussions. It reveals not just the topics of concern but also the depth of feelings associated with them, providing a richer, more human-centered perspective on economic narratives.

XIV. Comparative Sentiment Analysis by Income Level

Here we look at the sentiment behind objects, filtered by income level. It gives us an opportunity to see differences in opinion between these people.

Segmentation by Income Level

• Middle: Focus on economic stability and job-related concerns.

• Poor: Emphasis on living wages and control over economic conditions.

• Upper: Concerns about broader societal burdens and empathy towards lower earners.

• Working Class: Challenges around poverty alleviation and family support policies.

• Working Poor: Significant worries about contract workers and worker welfare.

XVV. Insights & Policy Implications

This analysis highlights varied emotional landscapes across different income levels in response to economic issues. Policymakers and social planners can use these insights to tailor interventions and communications that resonate with each group's specific concerns and emotional states.

Middle Income

• Number of Unique Objects: 14

• Most Common Emotion: Joy

• Average Emotion Score: 0.387

• Insight: Middle-income groups tend to have a more positive sentiment (joy) towards economic issues. The diversity in objects discussed indicates a range of concerns but with a generally optimistic outlook.

• Policy Implication: Policies could focus on sustaining and enhancing this group's economic stability, possibly through job security measures and support for middle-class financial pressures.

Poor

• Number of Unique Objects: 4

• Most Common Emotion: Anger

• Average Emotion Score: 0.409

• Insight: The prevalence of anger suggests significant frustration or dissatisfaction with economic conditions among the poor. The fewer number of unique objects might indicate concentrated concerns on specific issues.

• Policy Implication: Policies should prioritize addressing the core issues causing frustration, such as living wages, affordable housing, and social welfare support.

Upper Income

• Number of Unique Objects: 4

• Most Common Emotion: Sadness

• Average Emotion Score: 0.503

• Insight: Upper-income groups exhibit sadness, possibly reflecting empathy towards societal burdens or awareness of broader economic challenges.

• Policy Implication: There's an opportunity to engage this group in solutions for broader societal challenges, leveraging their resources and influence for societal benefit.

Working Class

• Number of Unique Objects: 28

• Most Common Emotion: Sadness

• Average Emotion Score: 0.467

• Insight: A wide range of economic concerns is reflected, with sadness being predominant. This suggests a sense of struggle or dissatisfaction with current economic circumstances.

• Policy Implication: Focus on policies that address the diverse needs of this group, including job security, wage fairness, and access to affordable services.

Working Poor

• Number of Unique Objects: 10

• Most Common Emotion: Joy

• Average Emotion Score: 0.407

• Insight: Despite economic challenges, there's a sense of resilience or positive sentiment in facing these challenges.

• Policy Implication: Supportive policies that bolster this resilience, such as skill development programs, financial literacy, and access to better-paying jobs, could be beneficial.

The best way to start that process isn’t just looking at keywords. It’s using the entirety of the text to show us what’s most important. 

To get underway, we ran a Latent Dirichlet allocation or LDA. An LDA is one of the most popular methods for performing topic modeling. 

Each document consists of various words and each topic can be associated with some words. The purpose of the LDA is to find topics that the document belongs to, on the basis of words in the text. It assumes that documents with similar topics will use a similar group of words; as you never find stories about sports teams containing text about baking apple pies. This enables the documents to map the probability distribution over latent topics and topics are probability distribution.

I. Building Topics with Latent Dirichlet Allocation

These detailed topic analyses reveal how different economic and social issues are perceived and discussed within the community. The emotional and sentiment undercurrents provide valuable insights into public mood and priorities, which are crucial for crafting effective narratives and communication strategies.

Topic 0: Employment and Living Conditions

• Top Keywords: Rent, Living, Wage, Leave, Time, Paid, Work, Loss, Money, Live

• Sentiment Profile: Balanced emotions with a tendency towards joy and sadness.

• Influence: This topic likely resonates with individuals facing challenges related to living wages, rent, and work-life balance. The mix of joy and sadness suggests a community that is hopeful yet burdened by current economic conditions.
  
  

Topic 1: Economic Growth and Income Inequality

• Top Keywords: Wage, Living, Income, People, 000, Create, Make, Increase, Minimum, Believe

• Sentiment Profile: Higher joy, indicating optimism, mixed with significant sadness.

• Influence: Discussions here revolve around income disparities and aspirations for economic growth. The optimism could reflect a belief in the potential for improvement, while the sadness might indicate current dissatisfaction with income inequality.
  
  

Topic 2: Social Justice and Housing

• Top Keywords: Control, Reparations, America, United, States, Housing, Month, Allow, Poverty, Lot

• Sentiment Profile: High levels of anger and sadness.

• Influence: This topic encompasses social justice issues, including housing and poverty. The strong presence of anger and sadness suggests deep-seated frustrations and concerns about inequality and social policies.
  
  

Topic 3: Urban Issues and Housing in Michigan

• Top Keywords: Basic, Housing, People, Experiencing, Income, Michigan, Issue, Want, Detroit, Affordable

• Sentiment Profile: Predominantly sadness, with a notable amount of fear.

• Influence: Focused on housing challenges, particularly in Michigan and Detroit. The prevailing sadness and fear may reflect concerns about housing affordability and the social implications of urban policies.
  
  

Topic 4: Social Support Systems

• Top Keywords: Childcare, Assistance, Control, Rent, Income, Expand, Insurance, Unemployment, Basic, Making

• Sentiment Profile: Moderate levels of sadness and low joy.

• Influence: This topic highlights issues related to social support systems like childcare and unemployment assistance. The sentiments indicate a community that is somewhat despondent about the current state of social support, yet engaged in seeking improvements.

Summary

These detailed topic analyses reveal how different economic and social issues are perceived and discussed within the community. The emotional and sentiment undercurrents provide valuable insights into public mood and priorities, which are crucial for crafting effective narratives and communication strategies.

Next, we can use this enriched understanding to further refine our audience profiles and develop more targeted narratives. If there's a specific aspect you'd like to focus on next, please let me know!

II. Informing Our Topics With Latent Semantic Analysis

This process provides a nuanced understanding of the emotional responses associated with each topic. These insights are invaluable for tailoring communication strategies to effectively resonate with the community's feelings and perspectives. The sentiment profiles guide us in creating narratives that not only address the topics of interest but also acknowledge and respond to the underlying emotional tones.

Topic 0: Employment and Living Conditions

• Dominant Sentiment: Balanced between joy and sadness.

• Anger: Moderate, possibly reflecting frustration with work conditions or living expenses.

• Fear: Notable, indicating concerns about job security or the impact of economic changes.

• Influence: The emotional landscape here suggests a community actively concerned with and impacted by employment and living conditions. The presence of both joy and sadness could indicate a community that is hopeful for change but currently struggling with the status quo.

Topic 1: Economic Growth and Income Inequality

• Dominant Sentiment: Optimism (Joy) mixed with significant sadness.

• Anger: Lower, suggesting less direct frustration or confrontation.

• Fear: Moderate, potentially reflecting concerns about the future and economic stability.

• Influence: The optimism in this topic could signify a belief in potential solutions to income inequality, while the sadness might indicate a recognition of the current challenges and disparities in economic growth.

Topic 2: Social Justice and Housing

• Dominant Sentiment: High levels of anger and sadness.

• Fear: Relatively lower, suggesting that concerns are more about current injustices rather than future threats.

• Influence: The strong emotions in this topic are indicative of deep concerns about social justice and housing issues. The high levels of anger and sadness highlight the urgency and intensity of these concerns within the community.

Topic 3: Urban Issues and Housing in Michigan

• Dominant Sentiment: Predominant sadness with fear.

• Anger: Lower, indicating perhaps a sense of resignation or helplessness.

• Influence: The emotions here point to a community deeply affected by urban housing issues, particularly in Michigan. The high sadness and fear could be reflecting concerns about affordability, displacement, and the broader impact of housing policies on communities.

Topic 4: Social Support Systems

• Dominant Sentiment: Moderate sadness and low joy.

• Anger: Lower, which might suggest a level of acceptance or the need for dialogue and solutions.

• Influence: The sentiment profile here indicates a community that is perhaps resigned to the challenges in social support systems but still engaged in seeking improvements. The moderate sadness could reflect dissatisfaction with current support mechanisms.

III. Creating a Better Content Plan

Now that our content topics are in alignment with the emotions of respondents and we have a battery of models to create with, let's map out a content strategy.

By thoughtfully integrating emotion and sentiment into your content across these platforms and media types, you create a powerful, multi-dimensional campaign that speaks directly to the hearts and minds of your audience. This strategy not only increases awareness and understanding of the issues but also drives engagement and mobilizes action towards positive change.

Remember, the success of such a campaign lies in its ability to consistently deliver content that resonates emotionally and intellectually with the audience, across all platforms. Tailoring the sentiment to fit the platform and audience preferences is key to maximizing impact and reach.

Facebook & WhatsApp

• Target: Broad age range, especially adults.

• Content: Share emotive stories like Alex's struggle with living wages, focusing on the emotional journey (hardships and hope). Use videos and posts to highlight these narratives with a mix of sadness, anger, and optimism. Include infographics that simplify complex policy issues.

• Engagement: Create discussion threads and groups for deeper engagement. Use emotional appeals to encourage sharing personal stories and opinions. Facilitate live sessions with experts or activists, using sentiment to drive meaningful conversations.

• Sentiment: Balance between highlighting struggles (sadness, anger) and offering solutions (hope, optimism).

Twitter

• Target: Influencers, policymakers, and activists.

• Content: Craft tweets with compelling, concise messaging. Use impactful statements from narratives, backed by relevant data. Share quick updates on policy developments, incorporating emotional cues to evoke urgency or triumph.

• Engagement: Engage in trending hashtags to increase visibility. Retweet and respond to users' reactions, fostering a community of shared sentiment.

• Sentiment: Use a mix of urgent (fear, anger) and positive (inspiration, optimism) sentiments to create a dynamic and engaging Twitter presence.

TikTok & Instagram

• Target: Younger audience, visually oriented.

• Content: Create short, visually appealing videos that capture the emotional essence of narratives. Use creative storytelling to depict the challenges and aspirations of characters like Jordan and Taylor.

• Engagement: Encourage user-generated content through challenges or hashtags. Respond to comments to build a community feel.

• Sentiment: Focus on inspiration and empowerment, using joy and surprise to engage viewers.

Blog Posts & Landing Pages

• Target: Individuals seeking in-depth information.

• Content: Write detailed articles and stories, integrating emotional arcs and sentiment analysis to make the content more relatable and persuasive.

• Engagement: Include calls-to-action for joining mailing lists or participating in initiatives. Use comment sections for discussions.

• Sentiment: Blend informative (neutral, trust) with emotive (hope, concern) content to keep readers engaged and informed.

Podcast

• Target: Audiences interested in regular, in-depth discussions.

• Content: Feature stories, interviews, and discussions on relevant topics, using the emotional undertones of each narrative to guide the conversation.

• Engagement: Invite listeners to submit questions or topics, creating an interactive experience.

• Sentiment: Balance informative content with emotional storytelling to keep listeners engaged and returning for more episodes.

Video Channel

• Target: Wide audience range; great for visual storytelling.

• Content: Produce a variety of content, from documentary-style pieces about community activists to informative videos on policy issues, using emotional storytelling to engage viewers.

• Engagement: Encourage comments and shares to foster a community. Host live video sessions for real-time engagement.

• Sentiment: Use a mix of serious (concern, urgency) and uplifting (hope, success) content to cater to a diverse audience.

Print & Outdoor Advertising

• Target: Local communities, broader public.

• Content: Design posters and flyers with compelling visuals and messages that encapsulate the emotional essence of your campaign.

• Engagement: Place these in high-traffic areas to spark curiosity and direct people to your digital platforms for more information.

• Sentiment: Use striking imagery and powerful words to evoke strong emotions (urgency, optimism) and prompt action.

Overall Campaign Sentiment Strategy

• Highlight Struggles & Challenges: Use emotions like sadness and anger to highlight the real challenges faced by individuals in the narratives. This approach fosters empathy and a sense of urgency.

• Inspire Action & Change: Balance the narrative with optimism and hope, showcasing success stories and potential solutions to motivate and engage the audience.

• Foster Community & Solidarity: Encourage sharing of personal stories and experiences, building a sense of community and shared purpose.

IV. Creating a Better Digital Ad Strategy

By thoughtfully combining emotion, sentiment, and contextual targeting in your Google Ads campaign, you can effectively drive traffic and engagement, aligning with your campaign's broader objectives. Remember, the key is to stay authentic and true to your campaign's core messages while meeting your audience where they are.

Understanding Emotion, Sentiment, and Context in Google Ads

• Emotion & Sentiment: Your ads should evoke emotions aligned with your campaign's goals. For instance, inspire action (optimism, hope) or highlight issues (concern, urgency).

• Context: Consider the user's search intent and how it aligns with your campaign. Use keywords and ad placements that match the context of your audience's interests and needs.

Educational Reform and Advocacy

• Keywords: Educational policy reform, community education initiatives, volunteer teaching opportunities.

• Emotion/Sentiment: Inspiration, empowerment.

• Ad Content: Highlight stories or initiatives like Jordan's narrative, emphasizing the transformative power of education.

Economic Justice and Living Wages

• Keywords: Living wage advocacy, economic justice movements, fair employment practices.

• Emotion/Sentiment: Urgency, solidarity.

• Ad Content: Focus on personal stories like Alex's, emphasizing the need for fair wages and better living conditions.

Social Justice and Housing Rights

• Keywords: Housing justice, urban development policies, affordable housing initiatives.

• Emotion/Sentiment: Concern, activism.

• Ad Content: Use narratives like Morgan's to highlight the importance of equitable housing policies and community activism.

Community Engagement and Local Issues (specific to regions like Michigan)

• Keywords: Community engagement Michigan, urban renewal Detroit, local activism.

• Emotion/Sentiment: Community spirit, determination.

• Ad Content: Share stories and initiatives relevant to local audiences, like Jordan's efforts in Detroit.

Support Systems and Social Welfare

• Keywords: Social support advocacy, childcare reform, unemployment assistance.

• Emotion/Sentiment: Compassion, support.

• Ad Content: Use narratives like Sam's to discuss the importance of robust social support systems.

Implementing the Strategy

• Targeting and Ad Placement: Use demographic and geographic targeting to reach the right audience. Consider where your audience spends time online and target those spaces.

• Landing Pages: Create landing pages that match the emotional tone of your ads. Ensure a seamless transition from the ad to the page, maintaining narrative consistency.

• Testing and Optimization: Regularly test different ad copies and formats to see what resonates best with your audience. Use A/B testing for headlines, descriptions, and calls-to-action.

• Tracking and Analytics: Monitor your campaign performance closely. Use Google Analytics to track conversions, click-through rates, and other relevant metrics to understand the impact of your ads.