Module 1: Determine what data to collect

Determine what data to collect

Hallie: Fascinating data insights

Helpful resources and tips

As a learner, you can choose to complete one or multiple courses in the Google Data Analytics Certificate program. However, to obtain the certificate, you must complete all eight courses in the program and submit all graded assignments. This reading describes what is required to obtain a certificate, as well as best practices for you to have a successful learning experience on Coursera.

Course completion to obtain a certificate

To receive your official Google Data Analytics Certificate, you must:

• Pass all graded assignments in all eight courses of the certificate program. Each graded assignment is part of a cumulative graded score for the course, and the passing grade for each course is 80%.

AND one of the following:

• Pay the course certificate fee,

• Be approved for Coursera Financial Aid, or

• Complete the certificate through an educational institution, employer, or agency that's sponsoring your participation.

Evaluate your current data analytics skills

If you’re a learner who has some experience with data analytics, you may wish to bypass this certificate program and proceed directly to a more advanced program such as the Google Advanced Data Analytics Certificate or the Google Business Intelligence Certificate. Determine whether you’re ready for one of these programs in an upcoming reading, Evaluate your current data analytics skills.

Healthy habits for successfully completing the certificate

Here are some best practices to help you successfully complete the courses in the program:

• Plan your time: Setting regular study times and following them each week helps make learning a part of your routine. Use a calendar or timetable to create a schedule, and list what you plan to do each day in order to set achievable goals. Find a space that allows you to focus.

• Work at your own pace: Everyone learns differently, so this program has been designed to let you work at your own pace. Although your personalized deadlines start when you enroll, feel free to progress through the program at the speed that works best for you. There is no penalty for late assignments; to earn your certificate, all you have to do is complete all of the work. If you miss two assessment deadlines in a row, or if you miss an assessment deadline by two weeks, you'll see a Reset deadlines option on the Grades page. Click it to switch to a new course schedule with updated deadlines. You can use this option as many times as you need.

• Be curious: If you find an idea that gets you excited, act on it! Ask questions, search for more details online, explore the links that interest you, and take notes on your discoveries. The steps you take to support your learning will advance your knowledge, create more opportunities for you in this high-growth field, and help you qualify for jobs.

• Follow along with instructors: Each instructional video with an onscreen software or programming demonstration has a corresponding step-by-step guide. This guide is a reading that outlines each step an instructor performs in the video. You’ll also find additional information in the guide that explains the purpose of the concepts covered, why a step is done a certain way, and tips for using the tool.

• Take notes: Notes will help you remember information in the future, especially as you’re preparing to enter a new job field. In addition, taking notes is an effective way to make connections between topics and gain a better understanding of those topics.

• Review exemplars: Exemplars are completed assignments that fully meet an activity's criteria; many activities in this program have exemplars that you can compare to your own work. Although there are often many different ways to complete an assignment, exemplars offer you guidance and inspiration about how to complete the activity.

• Build your career identity: Your career identity is the unique value you bring to the workforce. Watch this video to learn about the key components of career identity and write your own career identity statement. Discovering and defining your own career identity makes you better equipped to choose a career path that aligns with your strengths, values, and goals and is more effective in your chosen profession.

• Study together: It can be hard to find the time to progress through a Google Career Certificate with all life’s distractions. Register for Study Group – live study sessions hosted Monday through Thursday that offer you dedicated time and space to focus on your coursework alongside a community of your peers. Participation in Study Group is optional for our learners.

• Connect with other learners: If you have a question, chances are, you’re not alone. Reach out in the discussion forum to ask for help from other learners in this program. You can also visit Coursera’s private Google Data Analytics Community to expand your network, discuss career journeys, and share experiences. Check out the quick start guide.

• Update your profile: Update your profile on Coursera to include your photo, career goals, and more. When other learners find you in the discussion forums, they can click on your name to access your profile and get to know you better.

Documents and files for course activities

To complete certain activities in the program, you will need to use digital documents, spreadsheets, presentations, and/or labs. Data analytics professionals use these software tools to collaborate within their teams and organizations. If you need more information about using a particular type of document, refer to these resources:

• Microsoft Word: Help and learning: Microsoft Support page for Word

• Google Docs: Help Center page for Google Docs

• Microsoft Excel: Help and learning: Microsoft Support page for Excel

• Google Sheets: Help Center page for Google Sheets

• Microsoft PowerPoint: Help and learning: Microsoft Support page for PowerPoint

• How to use Google Slides: Help Center page for Google Slides

Module, course, and certificate glossaries

This program covers a lot of terms and concepts, some of which you may already know and some of which may be unfamiliar to you. To review terms and help you prepare for graded quizzes, refer to the following glossaries:

• Module glossaries: At the end of each module’s content, you can review a glossary of terms from that module. The module glossaries are not downloadable; however, all of the terms and definitions are included in the course and certificate glossaries, which are downloadable.

• Course glossaries: At the end of each course, you can access and download a glossary that covers all of the terms in that course.

• Certificate glossary: The certificate glossary includes all of the terms in the entire certificate program and is a helpful resource that you can reference throughout the program or at any time in the future.

Find the course and certificate glossaries using the course’s Resources tab. Access and download the certificate glossaries, then save them on your computer. To access the Data Analytics Certificate program glossary, click the following link and select Use Template.

Link to the glossary: Google Data Analytics Certificate glossary

OR

If you don’t have a Google account, download the glossary directly from the following attachment.

Data Analytics Certificate glossary

DOCX File

Course feedback

Providing feedback on videos, readings, and other materials is easy. With the resource open in your browser, locate the thumbs-up and thumbs-down symbols.

• Click thumbs-up for materials that are helpful.

• Click thumbs-down for materials that are not helpful.

If you want to flag a specific issue with an item, click the flag icon, select a category, and enter an explanation in the text box. This feedback goes to the course development team and isn’t visible to other learners. All feedback received helps to create even better certificate programs in the future.

For technical help on the Coursera platform, visit the Learner Help Center

Data collection in our world

Determine what data to collect

1.

Question 1

What are cookies?

Status: [object Object]

1 point

Programs that enable users to access websites

Types of malware that can damage computers

Pieces of code that store information about a website

Small files stored on computers that contain information about users

2.

Question 2

Fill in the blank: For data analytics projects, _____ data is typically preferred because users know it originated within the organization.

Status: [object Object]

1 point

multi-party

first-party

second-party

third-party

3.

Question 3

A grocery store chain purchases customer data from a credit card company. The grocer uses this data to identify its most loyal customers and offer them special promotions and discounts. What type of data is being used in this scenario?

Status: [object Object]

1 point

First-party

Multi-party

Second-party

Third-party

4.

Question 4

In data analytics, what term refers to all possible data values in a dataset?

Status: [object Object]

1 point

Representation

Source

Population

Sample

Discover data formats

Activity Overview

Now that you’ve learned about unstructured data and data formats, pause for a moment and think about unstructured data and how it compares to structured data. Most of the data being generated right now is unstructured, and being able to understand it empowers businesses to uncover hidden patterns; trends; and important information that may be inaccessible through traditional, structured data analysis. Answering the questions in this self-reflection will help reinforce what you’ve learned and enable you to make the most of all kinds of data throughout your career.

Scenario

Review the following scenario. Then complete the instructions.

In this self-reflection, you’ll explore and reflect on the nature of unstructured data by interacting with a crowd-sourced dataset on the website Quick, Draw!

Quick, Draw! is a game in which people create drawings. Quick! Draw has used these drawings to amass a dataset that contains millions of images separated into categories such as plants, animals, or vehicles. On the Quick, Draw! website, you can examine these drawings or play the game to add your own. These drawings help train computers to recognize objects with artificial intelligence.

In this self-reflection, you’ll explore the Quick! Draw website and create your own drawings. Then, you’ll reflect on Quick! Draw’s dataset and what you can deduce about the nature of unstructured versus structured data.

Step-By-Step Instructions

Follow the instructions to complete each step of the activity. Then answer the questions at the end of the activity before going to the next course item.

Step 1: Explore the Quick, Draw! website

1. Visit the Quick, Draw! website.

2. Click the drop down arrow featured next to Now visualizing: cloud.

3. Select a type of doodle to begin.

4. Click an image to get its details, like the number of doodles drawn. For example, there are more than 100,000 different drawings of elephants.

5. Scroll through the list and determine if there are any that don’t belong. If you find one that doesn’t appear to match the intended object, click on it and select Flag as inappropriate.

6. Explore other categories of drawings. Choose three that interest you and examine their doodles.

Optional Step 2: Explore Get the data

Explore further. Click Get the data to visit the GitHub page containing the entire dataset. As you become more familiar with data projects and start creating your own, you can return to this dataset and analyze it yourself.

Optional Step 3: Explore Play the game

Click Play the game to draw your own doodles and contribute to Quick, Draw!’s dataset.

Continue exploring structured data

1.

Question 1

What type of data is the height of a skyscraper?

Status: [object Object]

1 point

Discrete

Nominal

Qualitative

Continuous

2.

Question 2

In data analytics, what is the term for data that is generated from, and lives, outside of an organization?

Status: [object Object]

1 point

Peripheral

External

Outer

Internal

3.

Question 3

What are the key characteristics of unstructured data? Select all that apply.

Status: [object Object]

1 point

Fits neatly into rows and columns

Clearly identifiable construction

Unorganized

May have an internal structure

4.

Question 4

Fill in the blank: A data model is used to organize _____ and how they relate to one another.

Status: [object Object]

1 point

database structures

spreadsheet fields

data elements

data visualizations

Know the type of data you're working with

The power of multiple conditions

For data analysts, the real power of Boolean logic comes from being able to combine multiple conditions in a single statement. For example, if you wanted to filter for shoes that were grey or pink, and waterproof, you could construct a Boolean statement such as: “IF ((Color = "Grey") OR (Color = "Pink")) AND (Waterproof="True")

Notice that you can use parentheses to group your conditions together.

Key takeaways

Operators are symbols that name the operation or calculation to be performed. The operators AND, OR, and NOT can be used to write Boolean statements in programming languages. Whether you are doing a search for new shoes or applying this logic to queries, Boolean logic lets you create multiple conditions to filter your results. Now that you know a little more about Boolean logic, you can start using it!

Resources for more information

• Learn about who pioneered Boolean logic in this historical article: Origins of Boolean Algebra in the Logic of Classes.

• Find more information about using AND, OR, and NOT from these tips for searching with Boolean operators.

Data table components

Step-by-Step: Meet wide and long data

This reading outlines the steps the instructor performs in the following video, Meet wide and long data. In this video, the instructor presents wide and long data formats and discusses the types of questions each format can help you answer.

Keep this guide open as you watch the video. It can serve as a helpful reference if you need additional context or clarification while following the video steps. This is not a graded activity, but you can complete these steps to practice the skills demonstrated in the video.

What you’ll need

If you would like to access the spreadsheets the instructor uses in this video, select the link to a dataset to create a copy. If you don’t have a Google account, download the data directly from the attachments below.

Link to population datasets:

• Population, Latin, and Caribbean Countries, 2010–2019, wide format

• Population, Latin, and Caribbean Countries, 2010–2019, long format

OR

Download data:

Population, Latin, and Caribbean Countries, 2010–2019, wide format

XLSX File

Population, Latin, and Caribbean Countries, 2010–2019, long format

XLSX File

Example 1: Examine wide data

Wide data is a dataset in which every data subject has a single row with multiple columns to hold the values of various attributes of the subject. It is helpful for comparing specific attributes across different subjects.

1. Open the Population, Latin, and Caribbean Countries, 2010–2019, wide format spreadsheet.

2. Each row contains all population data for one country.

3. The population data for each year is contained in a column.

4. Find the annual population of Argentina in row 3.

5. In this wide format, you can quickly compare the annual population of Argentina to the annual populations of Antigua and Barbuda, Aruba, the Bahamas, or any other country.

Find the country with the highest population in 2010

1. Select column E, which contains each country’s 2010 population data.

2. Right-click column header E and choose Sort Z to A.

3. Notice that Brazil is now at the top of the list because it had the highest population in the year 2010.

Find the country with the lowest population in 2013

1. Select column H.

2. Right-click column header H and choose Sort A to Z.

3. Notice that the British Virgin Islands are now at the top because they had the lowest population of all countries in 2013.

Example 2: Examine long data

Long data is data in which each row represents one observation per subject, so each subject will be represented by multiple rows. This data format is useful for comparing changes over time or making other comparisons across subjects.

1. Open the Population, Latin, and Caribbean Countries, 2010–2019, long format spreadsheet.

2. Notice the data is no longer organized into columns by year. All of the years are now in one column.

3. Find Argentina’s population data in rows 12-21. Each row contains one year of Argentina’s population data.

Meet wide and long data

2. Click on the other .csv file. This opens a second tab with the second dataset.

3. Take a moment to check out each dataset.

4. Sort the data in each column by clicking on the horizontal bars to the right of each column name.

5. Click on the button that says 10 of 17 columns to change the columns that are visible in the table.

In the dropdown menu, there’s a checkmark next to the name of each column that appears in the table. Checking or unchecking one of these boxes will change what data is presented.

Congratulations! You’ve explored several ways to interact with the dataset. This will help you get familiar with the Kaggle interface. You can save the notebook you worked in for future reference. Coming up, you’ll learn more about other ways you can use Kaggle.

1.

Question 1

Fill in the blank: A data type is a specific kind of data _____ that tells what kind of value the data is.

attribute

frame

model

point

1 point

2.

Question 2

What are the key characteristics of a text, or string, data type? Select all that apply.

Sequence of characters and punctuation

Contains textual information

Only two possible values

Has numerical percentages

1 point

3.

Question 3

In a data table, where are fields contained?

Rows

Charts

Favorites

Columns

1 point

4.

Question 4

When using long data, each subject has data in multiple rows. This is because each row represents what?

One observation per subject

Multiple values

True or false data points

Data in different formats

1 point

Glossary terms from module 1

Terms and definitions for Course 3, Module 1

Agenda: A list of scheduled appointments

Audio file: Digitized audio storage usually in an MP3, AAC, or other compressed format

Boolean data: A data type with only two possible values, usually true or false

Continuous data: Data that is measured and can have almost any numeric value

Cookie: A small file stored on a computer that contains information about its users

Data element: A piece of information in a dataset 

Data model: A tool for organizing data elements and how they relate to one another

Digital photo: An electronic or computer-based image usually in BMP or JPG format

Discrete data: Data that is counted and has a limited number of values

External data: Data that lives, and is generated, outside of an organization

Field: A single piece of information from a row or column of a spreadsheet; in a data table, typically a column in the table

First-party data: Data collected by an individual or group using their own resources

Long data: A dataset in which each row is one time point per subject, so each subject has data in multiple rows

Nominal data: A type of qualitative data that is categorized without a set order

Ordinal data: Qualitative data with a set order or scale

Ownership: The aspect of data ethics that presumes individuals own the raw data they provide and have primary control over its usage, processing, and sharing 

Pixel: In digital imaging, a small area of illumination on a display screen that, when combined with other adjacent areas, forms a digital image 

Population: In data analytics, all possible data values in a dataset

Record: A collection of related data in a data table, usually synonymous with row

Sample: In data analytics, a segment of a population that is representative of the entire population

Second-party data: Data collected by a group directly from its audience and then sold 

Social media: Websites and applications through which users create and share content or participate in social networking

String data type: A sequence of characters and punctuation that contains textual information (Refer to Text data type)

Structured data: Data organized in a certain format such as rows and columns

Text data type: A sequence of characters and punctuation that contains textual information (also called string data type)

United States Census Bureau: An agency in the U.S. Department of Commerce that serves as the nation’s leading provider of quality data about its people and economy

Unstructured data: Data that is not organized in any easily identifiable manner

Video file: A collection of images, audio files, and other data usually encoded in a compressed format such as MP4, MV4, MOV, AVI, or FLV

Wide data: A dataset in which every data subject has a single row with multiple columns to hold the values of various attributes of the subject

M2: Data Responsibility

Introduction to bias, credibility, privacy, and ethics

Bias: From questions to conclusions

Biased and unbiased data

Understand bias in data

1.

Question 1

Fill in the blank: Bias is a _____ preference in favor of or against a person, group of people, or thing.

Status: [object Object]

1 point

conscious or subconscious

sensible or insensible

fair or unfair

standard or substandard

2.

Question 2

Which of the following are examples of sampling bias? Select all that apply.

Status: [object Object]

1 point

An election poll only interviews people with college degrees.

A teacher gives higher grades to essays written in their own writing style.

A clinical study includes three times more men than women.

A survey of students does not include homeschooled students.

3.

Question 3

What is the term for the tendency to search for or interpret information in a way that validates pre-existing beliefs?

Status: [object Object]

1 point

Interpretation bias

Observer bias

Confirmation bias

Sampling bias

4.

Question 4

Which of the following terms are also ways of describing observer bias? Select all that apply.

Status: [object Object]

1 point

Experimenter bias

Perception bias

Research bias

Spectator bias

Identify good data sources

What is "bad" data?

1.

Question 1

Fill in the blank: Data is considered _____ when it is accurate, complete, and unbiased information that has been vetted and proven fit for use.

comprehensive

original

reliable

current

1 point

2.

Question 2

Which of the following are usually good data sources? Select all that apply.

Academic papers

Governmental agency data

Vetted public datasets

Social media sites

1 point

3.

Question 3

To determine if a data source is cited, ask which of the following questions? Select all that apply.

Who created this dataset?

When was this data last refreshed?

Has this dataset been properly cleaned?

Is this dataset from a credible organization?

1 point

4.

Question 4

A junior data analyst learns that the dataset they have been given is six years old. After looking into this further, they also discover that the age of the data is making the information irrelevant to their project. What good data source principle have they used to evaluate the dataset?

Comprehensive

Reliable

Original

Current

1 point

Essential data ethics

Optional refresher: Alex and the importance of data ethics

Prioritize data privacy

Data anonymization

What is data anonymization?

You have been learning about the importance of privacy in data analytics. Now, it is time to talk about data anonymization and what types of data should be anonymized. Personally identifiable information, or PII, is information that can be used by itself or with other data to track down a person's identity. 

Data anonymization is the process of protecting people's private or sensitive data by eliminating that kind of information. Typically, data anonymization involves blanking, hashing, or masking personal information, often by using fixed-length codes to represent data columns, or hiding data with altered values.

Your role in data anonymization

Organizations have a responsibility to protect their data and the personal information that data might contain. As a data analyst, you might be expected to understand what data needs to be anonymized, but you generally wouldn't be responsible for the data anonymization itself. A rare exception might be if you work with a copy of the data for testing or development purposes. In this case, you could be required to anonymize the data before you work with it. 

What types of data should be anonymized?

Healthcare and financial data are two of the most sensitive types of data. These industries rely a lot on data anonymization techniques. After all, the stakes are very high. That’s why data in these two industries usually goes through de-identification, which is a process used to wipe data clean of all personally identifying information.

Data anonymization is used in just about every industry. That is why it is so important for data analysts to understand the basics. Here is a list of data that is often anonymized:

• Telephone numbers

• Names

• License plates and license numbers

• Social security numbers

• IP addresses

• Medical records

• Email addresses

• Photographs

• Account numbers

For some people, it just makes sense that this type of data should be anonymized. For others, we have to be very specific about what needs to be anonymized. Imagine a world where we all had access to each other’s addresses, account numbers, and other identifiable information. That would invade a lot of people’s privacy and make the world less safe. Data anonymization is one of the ways we can keep data private and secure!

The ethical use of data

1.

Question 1

What are data ethics?

Well-founded standards of right and wrong that dictate how data is collected, shared, and used

Established methods for ensuring data is clean, well-organized, and appropriate for a project

Approved strategies data professionals use to safeguard the privacy and security of a dataset

Long-standing techniques for confirming that data is always used to benefit society

1 point

2.

Question 2

What concept states that all data-processing activities and algorithms should be completely explainable and understood by the individual who provides their data?

Privacy

Ownership

Currency

Transaction transparency

1 point

3.

Question 3

A data analyst removes personally identifying information from a dataset. What task are they performing?

Data anonymization

Data sorting

Data collection

Data visualization

1 point

4.

Question 4

Before completing a survey, an individual acknowledges reading information about how and why the data they provide will be used. What is this concept called?

Discretion

Currency

Privacy

Consent

1 point

Features of open data

The open data debate

Just like data privacy, open data is a widely debated topic in today’s world. Data analysts think a lot about open data, and as a future data analyst, you need to understand the basics to be successful in your new role.

What is open data?

In data analytics, open data is part of data ethics, which has to do with using data ethically. Openness refers to free access, usage, and sharing of data. But for data to be considered open, it has to:

• Be available and accessible to the public as a complete dataset

• Be provided under terms that allow it to be reused and redistributed

• Allow universal participation so that anyone can use, reuse, and redistribute the data

Data can only be considered open when it meets all three of these standards. 

The open data debate: What data should be publicly available?

One of the biggest benefits of open data is that credible databases can be used more widely. Basically, this means that all of that good data can be leveraged, shared, and combined with other data. This could have a huge impact on scientific collaboration, research advances, analytical capacity, and decision-making. But it is important to think about the individuals being represented by the public, open data, too.

Third-party data is collected by an entity that doesn’t have a direct relationship with the data. You might remember learning about this type of data earlier. For example, third parties might collect information about visitors to a certain website. Doing this lets these third parties create audience profiles, which helps them better understand user behavior and target them with more effective advertising. 

Personal identifiable information (PII) is data that is reasonably likely to identify a person and make information known about them. It is important to keep this data safe. PII can include a person’s address, credit card information, social security number, medical records, and more.

Everyone wants to keep personal information about themselves private. Because third-party data is readily available, it is important to balance the openness of data with the privacy of individuals.

Steps for ethical data use

Resources for open data

Luckily for data analysts, there are lots of trustworthy resources available for open data. It is important to remember that even reputable data needs to be constantly evaluated, but these websites are a useful starting point:

1. U.S. government data site: Data.gov is one of the most comprehensive data sources in the US. This resource gives users the data and tools that they need to do research, and even helps them develop web and mobile applications and design data visualizations. 

2. U.S. Census Bureau: This open data source offers demographic information from federal, state, and local governments, and commercial entities in the U.S. too. 

3. Open Data Network: This data source has a really powerful search engine and advanced filters. Here, you can find data on topics like finance, public safety, infrastructure, and housing and development.

4. Google Cloud Public Datasets: There are a selection of public datasets available through the Google Cloud Public Dataset Program that you can find already loaded into BigQuery.  

5. Dataset Search: The Dataset Search is a search engine designed specifically for data sets; you can use this to search for specific data sets.

Tour the dataset explorer

Notice that the Data Explorer has three viewing options: Detail, Compact, and Column. For now, we’ll focus on the Detail tab. 

The description at the top of the Detail tab shows that the umbrellas.csv file contains data on all the umbrellas in the video game. Let’s check out the columns. Each column header has three items: 

• A small icon on the left that shows the data type

• The name of the column

• An icon with three bars that lets you sort the data if you click on it 

Below each column header is a box that contains a summary of the data. This lets you quickly get an idea of what’s in the dataset. For example, the summary for the Name column shows there are 67 unique values for the umbrella names. The summary for the DIY column shows that 9 of the umbrella recipes are DIY, or “do it yourself.” Take a moment to explore the summaries for the other columns.

And that completes our tour! That’s a lot of information. Feel free to go back and review.

Access a dataset

After you’ve explored a dataset, you can link it to a Kaggle notebook or download it to access it for your own use. Linking a dataset to a Kaggle notebook means you create a new notebook from the existing dataset so that it is available for you to use.

Find your own datasets

Now, you’ll get a chance to choose your own datasets to work with! Use the following steps to find datasets that interest you:

1. When you’re ready, click on the Data icon on the left to return to the Datasets landing page.

2. Note that datasets can exist in a variety of formats. If you want to make sure your dataset is in a .csv format, click on the Filter button on the right side of the Datasets search bar. Then, choose CSV from the menu. 

3. Find 2-3 datasets that you’re interested in exploring further. 

4. Create notebooks from them, download them, or check them out in the Data Explorer. Keep these datasets in mind for your upcoming reflection.

Link or download a dataset

Here are the options to create a notebook or download the dataset: 

• Create a Kaggle notebook: To link a dataset to a Kaggle notebook, you click on the New Notebook button in the dataset header. This will create a notebook in your Kaggle account that links to the dataset.

• Download the dataset: To download a copy of the dataset to your computer, click on the Download button in the dataset header at the top of the page. 

• Open the file in Google Sheets: To open a Google Sheets view of the file, click on the download icon at the top-right of the Data Explorer. You can then download the file.

1.

Question 1

Fill in the blank: Openness refers to _____ access, usage, and sharing of data.

Status: [object Object]

1 point

limited

disclosed

protected

free

2.

Question 2

What is the preferred method for open data to be made available?

Status: [object Object]

1 point

A convenient and modifiable internet download

A secure password-protected file

A compressed file format that keeps file size small

A print copy that is easily shared by anyone

3.

Question 3

What are the main benefits of open data? Select all that apply.

Status: [object Object]

1 point

Restricts data access to certain groups of people

Combines data from different fields of knowledge

Good data is more widely available

Increases the amount of data available for purchase

4.

Question 4

What are the key aspects of universal participation? Select all that apply.

Status: [object Object]

1 point

No one can place restrictions on data to discriminate against a person or group.

Certain groups of people must share their private data.

All corporations are allowed to sell open data.

Everyone must be able to use, reuse, and redistribute open data.

Glossary terms from module 2

Terms and definitions for Course 3, Module 2

Bad data source: A data source that is not reliable, original, comprehensive, current, and cited (ROCCC) 

Bias: A conscious or subconscious preference in favor of or against a person, group of people, or thing

Confirmation bias: The tendency to search for or interpret information in a way that confirms pre-existing beliefs

Consent: The aspect of data ethics that presumes an individual’s right to know how and why their personal data will be used before agreeing to provide it

Cookie: A small file stored on a computer that contains information about its users

Currency: The aspect of data ethics that presumes individuals should be aware of financial transactions resulting from the use of their personal data and the scale of those transactions

Data anonymization: The process of protecting people's private or sensitive data by eliminating identifying information

Data bias: When a preference in favor of or against a person, group of people, or thing systematically skews data analysis results in a certain direction

Data ethics: Well-founded standards of right and wrong that dictate how data is collected, shared, and used

Data interoperability: A key factor leading to the successful use of open data among companies and governments

Data privacy: Preserving a data subject’s information any time a data transaction occurs

Ethics: Well-founded standards of right and wrong that prescribe what humans ought to do, usually in terms of rights, obligations, benefits to society, fairness, or specific virtues

Experimenter bias: The tendency for different people to observe things differently (also called observer bias)

Fairness: A quality of data analysis that does not create or reinforce bias 

First-party data: Data collected by an individual or group using their own resources

General Data Protection Regulation of the European Union (GDPR): Policy-making body in the European Union created to help protect people and their data

Good data source: A data source that is reliable, original, comprehensive, current, and cited (ROCCC) 

Interpretation bias: The tendency to interpret ambiguous situations in a positive or negative way

Observer bias: The tendency for different people to observe things differently (also called experimenter bias)

Open data: Data that is available to the public

Openness: The aspect of data ethics that promotes the free access, usage, and sharing of data

Sampling bias: Overrepresenting or underrepresenting certain members of a population as a result of working with a sample that is not representative of the population as a whole

Transaction transparency: The aspect of data ethics that presumes all data-processing activities and algorithms should be explainable and understood by the individual who provides the data

Unbiased sampling: When the sample of the population being measured is representative of the population as a whole

M3: Database essentials 

All about databases

Database features and components

You can use this data to determine whether a specific date falls on a weekend or holiday and add a column to your sheet that reflects this information. Then, you can find out whether sales on the weekends and holidays are greater than sales on other days. This will be useful to know for inventory planning and marketing purposes. 

Question 4: How does profitability differ for new customers versus returning customers?

Your dataset does not contain sales data related to new customers. Without this data, you won’t be able to answer your final question. However, it may be the case that the company collects customer data and stores it in a different data table. 

If so, your next step would be to find out how to access the company’s customer data. You can then join the revenue sales data to the customer data table to categorize each sale as from a new or returning customer and analyze the difference in profitability between the two sets of customers. This information will help your stakeholders develop marketing campaigns for specific types of customers to increase brand loyalty and overall profitability. 

Key takeaways

When working on analytics projects, you won’t always have all the necessary or relevant data at your disposal.  In many of these cases, you can turn to other data sources to fill in the gaps. 

 Despite the limitations of your dataset, it’s still possible to offer your stakeholders some valuable insights. For next steps, your best plan of action will be to take the initiative to ask questions, identify other relevant datasets, or do some research on your own.  No matter what data you’re working with, carefully inspecting your data makes a big impact on the overall quality of your analysis.

1.

Question 1

Fill in the blank: A relational database contains a series of _____ that can be connected to form relationships.

Status: [object Object]

1 point

tables

cells

fields

spreadsheets

2.

Question 2

What is the term for an identifier that references a database column in which each value is unique?

Status: [object Object]

1 point

Primary key

Foreign key

Relation

Field

3.

Question 3

What process do data professionals use to eliminate data redundancy, increase data integrity, and reduce complexity in a database?

Status: [object Object]

1 point

Composition

Normalization

Iteration

Manipulation

4.

Question 4

Fill in the blank: When using a relational database, data analysts write _____ to request data from the related tables.

Status: [object Object]

1 point

programs

queries

relationships

keys

Demystify metadata

Spreadsheets and electronically created documents

Spreadsheets and documents are already filled with a considerable amount of data, so it’s no surprise that they also include metadata such as title, author, creation date, number of pages, and user comments. Additionally, spreadsheet metadata includes tab names, tables, and columns. In the following example, the image demonstrates the metadata for an electronically created Google Sheet:

Key takeaways

Metadata can be found in photos, emails, spreadsheets, websites, and much more! In your daily life, you use metadata to stay organized. As a data analyst, you’ll use metadata to understand the content and context of your data, as well as how it’s structured. Metadata provides data analysts with information about a data’s file type, title, geolocation, who created it, who last modified it, and who has access to it. As a data analyst, it’s important to keep accurate records of metadata to ensure that you are able to find, use, preserve, and reuse data in the future. Remember, it will be your responsibility to manage and make use of data in its entirety; metadata is as important as the data itself.

Metadata and metadata repositories

As you’re learning, metadata is data about data. It clearly describes how and when data was collected and how it’s organized. Metadata puts data into context and makes the data more understandable. This helps data analysts use data to solve problems and make informed business decisions. 

In this reading, you’ll learn more about the benefits of metadata, metadata repositories, and metadata of external databases.

The benefits of metadata

Reliability

Data analysts use reliable and high-quality data to identify the root causes of any problems that might occur during analysis and to improve their results. If the data being used to solve a problem or to make a data-driven decision is unreliable, there’s a good chance the results will be unreliable as well.

Metadata helps data analysts confirm their data is reliable by making sure it is:

• Accurate

• Precise

• Relevant

• Timely

It does this by helping analysts ensure that they’re working with the right data and that the data is described correctly. For example, a data analyst completing a project with data from 2022 can use metadata to easily determine if they should use data from a particular file.  

Consistency

Data analysts thrive on consistency and aim for uniformity in their data and databases,  and metadata helps make this possible. For example, to use survey data from two different sources, data analysts use metadata to make sure the same collection methods were applied in the survey so that both datasets can be compared reliably. 

When a database is consistent, it’s easier to discover relationships between the data inside the database and data that exists elsewhere. When data is uniform, it is:

• Organized: Data analysts can easily find tables and files, monitor the creation and alteration of assets, and store metadata. 

• Classified: Data analysts can categorize data when it follows a consistent format, which is beneficial in cleaning and processing data.

• Stored: Consistent and uniform data can be efficiently stored in various data repositories. This streamlines storage management tasks such as managing a database.   

• Accessed: Users, applications, and systems can efficiently locate and use data.

Together, these benefits empower data analysts to effectively analyze and interpret their data.

Metadata repositories

Metadata repositories help data analysts ensure their data is reliable and consistent.

Metadata repositories are specialized databases specifically created to store and manage metadata. They can be kept in a physical location or a virtual environment—like data that exists in the cloud.

Metadata repositories describe where the metadata came from and store that data in an accessible form with a common structure. This provides data analysts with quick and easy access to the data. If data analysts didn’t use a metadata repository, they would have to select each file to look up its information and compare the data manually, which would waste a lot of time and effort.

Data analysts also use metadata repositories to bring together multiple sources for data analysis. Metadata repositories do this by describing the state and location of the data, the structure of the tables inside the data, and who accessed the user logs.

Metadata of external databases

Data analysts use both second-party and third-party data to gain valuable insights and make strategic, data-driven decisions. Second-party data is data that’s collected by a group directly from the group’s audience and then sold. Third-party data is provided by outside sources that didn’t collect it directly. The providers of this data are not its original collectors and do not have a direct relationship with any individuals to whom the data belongs. The outside providers get the data from websites or other programs that pull it from the various platforms where it was originally generated. 

Data analysts should understand the metadata of external databases to confirm that it is consistent and reliable. In some cases, they should also contact the owner of the third-party data to confirm that it is accessible and available for purchase. Confirming that the data is reliable and that the proper permissions to use it have been obtained are best practices when using data that comes from another organization.

Key takeaways

Metadata helps data analysts make data-driven decisions more quickly and efficiently. It also ensures that data and databases are reliable and consistent. 

Metadata repositories are used to store metadata—including data from second-party and third-party companies. These repositories describe the state and location of the metadata, the structure of the tables inside it, and who has accessed the repository. Data analysts use metadata repositories to ensure that they use the right data appropriately.

Manage data with metadata

Fun with metadata

1.

Question 1

A large company has several databases across its many departments. What kind of metadata describes how many locations contain a certain piece of data?

Representative

Administrative

Descriptive

Structural

1 point

2.

Question 2

A large metropolitan high school gives each of its students an ID number to differentiate them in its database. What kind of metadata are the ID numbers?

Representative

Administrative

Descriptive

Structural

1 point

3.

Question 3

An international nonprofit organization wants to merge third-party data with its own data. Which of the following actions will help make this process successful? Select all that apply.

Use metadata to standardize the datasets.

Use metadata to evaluate the third-party data’s quality and credibility.

Replace the incoming data’s metadata with its own company metadata.

Alter the internal metadata to more closely reflect the incoming metadata.

1 point

4.

Question 4

Fill in the blank: Data _____ is a process data professionals use to ensure the formal management of their organization’s data assets.

storage

sourcing

governance

organization

1 point

So many places to find data

Working with .csv files

In an earlier lesson, you explored comma separated values (.csv) files and how to import them into a Google Sheet or download them from a website. Data analysts use .csv files often, so throughout this course you will use .csv files to transfer data into data analysis programs for further analysis and visualization. .csv files are plain text files with an organized table structure that includes rows and columns. The values in each row are separated by commas. The table structure makes .csv files easy to understand, edit, manipulate, and use for data analysis.

A major advantage of .csv files is their widespread compatibility. They can be imported and exported by a vast range of data analysis tools and software programs.

Download .csv files

To use .csv files and upload them to data analysis programs you will first need to download them to your local device. Downloading a .csv file from a website can vary depending on your operating system or internet browser. Here are some ways you can download a .csv file:

• Click the download link or .csv attachment: Locate the link for the .csv file or attachment on the website. Click on it, and the download process will start.

• Right-click and Save: Right-click on the data table or element containing the .csv data. Choose Save as… or a similar option. Name the file and make sure the extension on the file is “.csv”.

• Force download: You can use the Alt key on your keyboard while clicking the link. This will trigger the download, and you will be able to find the .csv file in your Downloads folder.

Note: When using the Chrome browser or ChromeOS, .csv files may open in a new tab instead of downloading to your machine. If this happens, follow these instructions: 

• Select File from the menu bar, then select Save as Google Sheets. This will open the .csv file as a Google Sheet. 

• Select File from the menu bar, then select Download from the dropdown menu, then select Comma Separated Values (.csv).

Upload .csv files

You will often need to upload .csv files during the data analysis process. Here is how you do this:

• Locate the upload option: Each data analysis platform will have a designated button, menu option, or drag and drop area labeled Upload or Import. This is where you will upload your .csv file. 

• Choose your .csv file: Click Upload or Import on the platform you are using to open your file explorer. Select your .csv file. If you just downloaded a .csv file from the web, it will be located in your computer’s Downloads folder. 

• Initiate the upload: Once you've selected your .csv file, click Upload or Import.The platform may display a progress bar or message indicating that the upload is complete.

Note: Some platforms have restrictions on the file size or format of .csv files. Make sure your .csv files adhere to these requirements before uploading.

Key takeaways

Data analysis programs help us extract insights and knowledge from data. Using .csv files is essential in data analysis. Understanding how to easily download data from the web or add your data to these programs will allow you to complete data cleaning, visualizations, analysis, and so much more!

Import data from spreadsheets and databases

Import data dynamically

As you’ve learned, you can import data from some data sources, like .csv files into a Google spreadsheet from the File menu. Keep in mind that, when you use this method, data that is updated in the .csv will not automatically be updated in the Google Sheet. Instead, it will need to be manually—and continually—updated in the Google Sheet. In some situations, such as when you want to be able to keep track of changes you’ve made, this method is ideal. In other situations, you might need to keep the data the same in both places, and using data that doesn’t update automatically can be time-consuming and tedious. Further, trying to maintain the same dataset in multiple places can cause errors later on.

Fortunately, there are tools to help you automate data imports so you don’t need to continually update the data in your current spreadsheet. Take a small general store as an example. The store has three cash registers handled by three clerks. At the end of each day, the owner wants to determine the total sales and the amount of cash in each register. Each clerk is responsible for counting their money and entering their sales total into a spreadsheet. The owner has the spreadsheets set up to import each clerks’ data into another spreadsheet, where it automates and calculates the total sales for all three registers. Without this automation, each clerk would have to take turns entering their data into the owner’s spreadsheet. This is an example of a dynamic method of importing data, which saves the owner and clerks time and energy. When data is dynamic, it is interactive and automatically changes and updates over time.

In the following sections you’ll learn how to import data into Google Sheets dynamically.

IMPORT functions in Google Sheets

The IMPORTRANGE function

In Google Sheets, the IMPORTRANGE function can import all or part of a dataset from another Google Sheet.

To use this function, you need two pieces of information:

1. The URL of the Google Sheet from which you’ll import data.

2. The name of the sheet and the range of cells you want to import into your Google Sheet.

Once you have this information, open the Google Sheet into which you want to import data and select the cell into which the first cell of data should be copied. Enter = to indicate you will enter a function, then complete the IMPORTRANGE function with the URL and range you identified in the following manner: =IMPORTRANGE("URL", "sheet_name!cell_range"). Note that an exclamation point separates the sheet name and the cell range in the second part of this function.

An example of this function is:

=IMPORTRANGE("https://docs.google.com/thisisatestabc123", "sheet1!A1:F13")

Note: This URL is for syntax purposes only. It is not meant to be entered into your own spreadsheet.

Once you’ve completed the function, a box will pop up to prompt you to allow access to the Google Sheet from which you’re importing data. You must allow access to the spreadsheet containing the data the first time you import it into Google Sheets. Replace it with a spreadsheet’s URL that you have created so you can control access by selecting the Allow access button.

Refer to the Google Help Center's IMPORTRANGE page for more information about the syntax. You’ll also learn more about this later in the program.

The IMPORTHTML function

Importing HTML tables is a basic method to extract data from public web pages. This process is often called “scraping.” Web scraping made easy introduces how to do this with Google Sheets or Microsoft Excel.

In Google Sheets, you can use the IMPORTHTML function to import the data from an HTML table (or list) on a web page. This function is similar to the IMPORTRANGE function. Refer to the Google Help Center's IMPORTHTML page for more information about the syntax.

The IMPORTDATA function

Sometimes data displayed on the web is in the form of a comma- or tab-delimited file.

You can use the IMPORTDATA function in a Google Sheet to import data into a Google Sheet. This function is similar to the IMPORTRANGE function. Refer to Google Help Center's IMPORTDATA page for more information and the syntax.

Explore public datasets

Open data helps create a lot of public datasets that you can access to make data-driven decisions. Here are some resources you can use to start searching for public datasets on your own:

• The Google Cloud Public Datasets allow data analysts access to high-demand public datasets, and make it easy to uncover insights in the cloud. 

• The Dataset Search can help you find available datasets online with keyword searches. 

• Kaggle has an Open Data search function that can help you find datasets to practice with.

• Finally, BigQuery hosts 150+ public datasets you can access and use. 

Public health datasets

1. Global Health Observatory data: You can search for datasets from this page or explore featured data collections from the World Health Organization.  

2. The Cancer Imaging Archive (TCIA) dataset: Just like the earlier dataset, this data is hosted by the Google Cloud Public Datasets and can be uploaded to BigQuery.

3. 1000 Genomes: This is another dataset from the Google Cloud Public resources that can be uploaded to BigQuery. 

Public climate datasets

1. National Climatic Data Center: The NCDC Quick Links page has a selection of datasets you can explore. 

2. NOAA Public Dataset Gallery: The NOAA Public Dataset Gallery contains a searchable collection of public datasets.

Public social-political datasets

1. UNICEF State of the World’s Children: This dataset from UNICEF includes a collection of tables that can be downloaded.

2. CPS Labor Force Statistics: This page contains links to several available datasets that you can explore.

3. The Stanford Open Policing Project: This dataset can be downloaded as a .csv file for your own use.

1.

Question 1

What are some key benefits of open-data initiatives? Select all that apply.

Status: [object Object]

1 point

Help educate citizens about important issues

Limit opportunities for collaboration

Make government activities more transparent

Support innovation and economic growth

2.

Question 2

What type of file saves data in a table format?

Status: [object Object]

1 point

Compatible scientific variables (.csv)

Comma-separated values (.csv)

Calculated spreadsheet values (.csv)

Cell-structured variables (.csv)

3.

Question 3

Bringing data from a .csv file into a spreadsheet is an example of what process?

Status: [object Object]

1 point

Filing data

Editing data

Normalizing data

Importing data

4.

Question 4

In Google Sheets, what function enables a data analyst to specify a range of cells in one spreadsheet to be duplicated in another?

Status: [object Object]

1 point

SPECIFY

CELLRANGE

DUPLICATE

IMPORTRANGE

Sort and filter to focus on relevant data

 Clean data in spreadsheets with sorting and filtering

8. Once both selections have been made, click Sort.

Now, If you scroll through the data, you’ll notice that the age range of the students at Gabriel Pereira (GP) is 15-22 years, and the age range of the students at Mouzinho da Silveira (MS) is 15-20 years. It appears that both schools have similar age ranges, but the GP school has students that are a little older.

By sorting the data, you’ve discovered a potential problem with the data. Because this dataset represents high school student achievement, any age older than 18 may indicate that a mistake was made when entering that student's age. You now know what age data may need to be researched and corrected. Your next step is to ask the superintendent about the legitimate age range for students in public high school. Then, you’ll know what age data is incorrect and should be removed.

Step 4: Filter data

The superintendent tells you that the maximum age limit for which public education is provided is 19 years old and that the age range should be 15-19 for both schools. Any student outside this age range should be deleted from the dataset. 

To clean your data, you need to remove the ages 20, 21, and 22 from your dataset. You can start by applying a filter to the age column. Filtering is the process of showing only the data that meets a specified criteria while hiding the rest. Filtering makes it easier to find data that you need.

1. First, apply a filter to the age column. Select the age column by clicking the letter at the top of the column (C). 

2. Then, from the menu bar, select Data, then Create a filter.

3. You can now inspect the values in the age column by going to the top of the column and clicking the Filter icon (). 

4. In Google Sheets, there are nine possible values for the field (15, 16, 17, 18, 19, 20, 21, and 22). You may notice that all the values have check marks. Filter this column for the values you want to select by unchecking all the other values (15, 16, 17, 18, and 19).

5. Then, click OK. This will single out the rows that contain the ages 20, 21, and 22. After you apply the filter, there should be nine such rows (seven for the GP school and two for the MS school). 

6. To delete the nine rows, first select them by clicking their row numbers. 

7. Then, from the menu bar, select Edit and Delete selected rows.

8. Click the Filter icon at the top of the age column to inspect the values once again. Now that you’ve removed the three incorrect ages (20, 21, and 22), there are five ages remaining (15, 16, 17, 18, and 19). The remaining ages are legitimate and can be used for analysis.

9. Finally, turn off the filter. From the menu bar, select Data and Turn off filter.

Step 5: Fill In Missing Data

Filling in missing data is an important part of data cleaning. It’s your job to fill in these blank spaces in your data with accurate values.

The superintendent wants to know what factors influence student performance, and a student’s reason for choosing a specific school will be important to know for analysis. The reason column shows the main reason a student chose to enroll in a specific school, according to their survey response: for example, because of the school’s reputation, or because it offers certain courses,  etc. So, you need to make sure the reason column is complete and without blanks. 

1. Start by applying a filter across the entire spreadsheet. Click on any cell in the sheet. Then, from the menu bar, select Data and Create a filter.

2. All the cells are now highlighted, and there are filters at the top of every column containing data. Click the Filter icon on the reason column (K). 

3. You may notice that the data values in the reason column include blanks. Filter this column for blanks by unchecking all the other values (course, home, reputation).

4. Then, click OK. Now, your sheet shows all the blank rows in the reason column. 

5. To clean your data, you need to find a good way to fill in these missing values. In this case, you cannot know what each missing value should be (that is, without a new survey, you can’t discover each student’s reason for choosing a specific school). So, you can replace the missing values with the value none_given. To do this while the column is still filtered for blanks, enter none_given in the first empty cell (K38). Then, press Enter.

6. Select cell K38 again. A small blue square, known as the fill handle, appears in the bottom-right corner of the cell. Double click the fill handle to fill all the other blank cells with the value none_given.

7. Finally, turn off the filter. From the menu bar, select Data and Turn off filter. If you scroll down the reason column, you should find that the value none_given has replaced all the blanks in the reason column.

Step 6: Convert Data

During the data analysis process, it's sometimes necessary to change text data (words) to numeric data (numbers). For example, some statistical packages like those used to perform machine learning will only accept numeric data values as input. 

In this case, the superintendent wants to know if a parent’s education level is a significant factor in student performance. The relevant data is in the Medu and Fedu columns--which, respectively, refer to the level of education of a student’s mother and father. Currently, the data is in text format. For the purposes of analysis, it will be useful to know the average education level of each student’s parents. To make this calculation, you first need to convert the data in the Medu and Fedu columns to number format.

To do this, you can match specific number values to the text data in each column. Start with the Medu column. If you click on the Filter icon at the top of the Medu column (G), you’ll notice the column contains the text data shown in the table below. You can use the following numeric codes for each piece of text data:

5. Then, click Replace all. 

6. While still in the popup window, repeat this process (steps 4-5) for the other four educational levels: primary education (4th grade), 5th to 9th grade, secondary education, and higher education. 

7. After replacing all five educational levels with numeric values, click Done to close the pop-up window. 

8. Check out your spreadsheet. All the cells in the Medu column now display numeric values.

9. Change the text data in the Fedu column (H) in the same way.

Pro-Tip: Save the activity template

Be sure to save a copy of the spreadsheet template you used to complete this activity. You can use it for further practice or to help you work through your thought processes for similar tasks in a future data analyst role.