Doing a survey for your IA can be a really good choice. It can also be a disaster. You have to think long and hard about what you want to find out and the questions you will ask. The pay off is the collection of data is relatively simple and you don't even have to leave your computer? If you want to do a survey and score highly the first thing to do is have a read through Cami's. Her IA scored 27/30 and is brilliant!

to get a 7 you must use both descriptive and inferential statistics - this is a long video but will really improve your understanding of how to analyse your statistics

level of measurement

1. category - male/female

2. ordinal - similar but there is an order between them like age categories

3. continuous - any question that could be answered with a number. How many times did you buy a plastic bottle this week.

(The choice of question will dictate what analysis you can do)

Step 1: Allocate numbers to your questions (weight)

Step 2: Transfer the info to a spreadsheet (easily done in Google forms)

Step 3: Analyse - what is the RQ - you need to answer it/

Step 4: Count and determine frequencies in % and bar and pie charts

Step 4: Ordinal questions

Step 5: Measures of mean and standard deviation - how often? What is the mean? You an use the range or use standard deviation.

Step 6: Bi variant analysis - look at pairs of questions - you can cross tabulate 2 questions e.g gender and age - you can then do a stacked bar chart but it is only a summary of the data

Step 7: You can use a scatter plot two continuous questions and find a positive or negative relationship - is there a relationship

RAISING RIGOR

Step 8: Can we use the data to make inferences. This is called inferential statistics to look at the significance of the data

Step 9: Level of confidence is 95%

Step 10: Category type questions - is one category statistically larger than the other = 1 way Chi squared test. Were there more male or females?

Step 10: 2 categories - is the difference down to chance? 2 way Chi squared test?

Step 11: Compare means - is it statistically significant - you have to use a T test -

Step 12: Scatterplot - use of one thing against another; Use a correlation test. Is it significant?

Excellent sentence starters for you results, analysis and conclusion section

Phrases to Use

This study found that...

The results should be interpreted with caution...

Findings from this study should be considered in light of several limitations.

Many/several variables that could contribute to this research were not accounted for as...

The data supports / does not support the hypothesis because...

Although the data shows a correlation / indicates support for...

There is strong / weak support for the conclusion as...

The data supports a conclusion that

More data is needed to provide support for this initial conclusion.

The reliability of the data is high / low due as shown by the low / high standard deviation of each mean.

Supporting Conclusions with Statistics

The standard deviation bars overlap and therefore there is too much variation in the data to draw a strong conclusion.

The standard deviation bars do not overlap suggesting confidence in the reliability of the data.

The correlation between the two variables is strong with an R² value approaching 1.

The correlation between the two variables is weak / does not exist as the R² value is low, approaching 0.

The t-test at 95% confidence supports the hypothesis that there is a significant difference between the two [locations].

There is a negative / positive correlation between x and y. As the x becomes [warmer] the y decreases. This is supported by a Pearson coefficient R² value of 0.3 which indicates a weak correlation, however with further statistical analysis, using the Spearman Rank coefficient

Presenting Data

When presenting your data you should introduce the data, add the table and the title / figure number, and then talk about the data, using the phrase, as seen in Fig. x, the data indicates that...

One of the decisions you will have to make is whether to include your raw data or just the processed data (calculated means and standard deviations for example). In investigations and extended essays where there is a lot of data then it is usually appropriate to only include the processed data e.g. survey responses but when the data is easily fitted into one page then it would be appropriate to include the raw data. Raw data should be included in an appendix if not included in the body of the report.

Another consideration is whether to include data in a table or a graph. Sometimes it is unnecessary duplication to do both. Ask the question, what is the purpose of this representation of the data. Does a graph aid understanding? Bar charts with error bars, for example, are easier to interpret than a table of the same data. A scatter graph with the addition of a trendline and R² value will show a correlation (or not) much more clearly than a table of the same data but in this case, the data in table form provides more precise information about particular points.

You should use the following 4 subheadings for this section.

1. Evaluation of conclusion

   • start with the research question and answer it clearly - "the evidence presented in this investigation support/do not support the hypothesis. There is evidence to suggest a there is a correlation between x and y. AND - how does is it link to the environmental issue - what light does it throw onto the subject

2. Evaluation of method; You must mention strengths and weaknesses of method and limitations

3. Modifications to method; show how you could address each weakness with an improvement to the method

4. Further research: Explain what else could be studied and what research should be done to improve the understanding of the EI more

There are 3 aspects in this section

1. evaluates the conclusion in the context of the environmental issue

2. discusses strengths, weaknesses and limitations within the method used

3. suggests modifications addressing one or more significant weaknesses with large effect and further areas of research.

1. What is the examiner looking for in the evaluation?

• evaluation of the conclusion with respect to the RQ and EI

• Discussion: strengths, weaknesses and limitations of the method

• Evaluation of the investigation

• proposed modifications

You have to discuss and evaluate the conclusion separately from stating them in the RAC. You must start with the original RQ then see if secondary data/research to support/refutes the conclusion. Is there evidence to suggest the conclusion is in agreement with what else you have read. Is the conclusion surprising? Are there reasons why?

2. Strengths and weaknesses and limitations

What is the examiner looking for in the evaluation?

• a discussion of strengths - this might be general or refer to specific parts that worked well.

• discussion of the reliability of the data

• identification of weaknesses in the method

• the evaluation of the relative impact of a weakness in the investigation

If you only address practical issues by saying just carrying out the procedure better, then you can only get a maximum 4. If you only give one element of strengths, weaknesses and limitations is completely missed out then the maximum is a 4 in this aspect. So give strengths, weaknesses and limitations.

3. Modifications and further research

What is the examiner looking for in the modifications and further research section?

• suggestions of modifications

• Focus on one or two significant issues

• Further areas of research

Suggestions modifications/further research should be precise, focused and relevant to the investigation. You must discuss how the modifications you suggest might bring the experimental results closer to what is expected.

so;

1. suggest a modification that is related to one of the weaknesses you have identified

2. then say how this modification will contribute to the reliability, precision and accuracy of the results

3. do this for two modifications

The further research suggested should follow on from the research in a meaningful way and go beyond the original method of investigation to show how it will enhance understanding of the EI or RQ

Here is what the examiner said about the Malaysian stream investigation; this only scored a 3 so the weakest part of the investigation.

" Discussion is more a link to, a continuation or repetition of the CXT. Very little connection between the conclusion and the EI. No strengths, weaknesses/limitations described. Suggests modifications and further areas of research. No strengths so max 4 in that band; in 1-2 for modifications, and further research is weak. Overall a 3"

You must discuss a reasonable solution to the problem. Your solution must include strengths and weaknesses/limitations of the solution.

BE very careful about generalisations- you have created a survey for Aiglon students and you cannot assume the results of your survey are reflective of the whole school population, the canton, the country or the world. Be very clear as you try to articulate this notion.

Do not make sweeping generalisations but indicate why your survey needs more research