1 of 20

Analyzing Data

Module 11

1

2 of 20

Introduction

2

3 of 20

By the end of this module, you will be able to...

Become familiar with what analyzing data is and�
Identify the main techniques to analyze quantitative and qualitative data.

3

4 of 20

Setting the Stage

This is the FUN part!

Simple is good.

It is a journey, not destination.

Partnering.

4

5 of 20

Getting started

Make a copy of the original data set
Prepare the data
“Clean” the data
Remember what your original questions were

5

It is important to make a copy of your original data set. This would let you work on your analyses, but still have the original data in case you need to go back and compare.

Prepare your data for analysis by organizing it. This may mean gathering all of the documents in one location. This may also include inputting data into a spreadsheet.

For most of the data you will need to check for errors. This is called “cleaning the data.” For example, check that the audience answered all the questions, or that the answer is in the acceptable range. Sometimes you can correct the error but in some other cases you may have to treat the error as a missing data. If you want to compare pre to post data, you may need to match the tests so that you can compare each individual person’s responses.

Then, you should look back at your evaluation questions. What did you want to learn from your data, in the first place? Your analysis should be done in a way that will help you to answer those questions.

Then, you are ready to start with your analysis!

6 of 20

Quantitative Data Analysis

6

(Diamond, 2016)

7 of 20

Tables

7

(Diamond, 2016)

8 of 20

Graphs

Bar graph

Scatter plot

Pie chart

8

(Diamond, 2016)

9 of 20

Statistical Techniques

9

(Diamond, 2016)

10 of 20

Qualitative Data Analysis

Descriptive text

Direct quotes

Verbal descriptions

Photographs

Drawings

10

(Diamond, 2016)

11 of 20

Data Reduction

11

(Diamond, 2016)

The first step for analyzing qualitative data is to reduce the data. You won’t be able to show a whole interview transcript so how can you show just what was important from that transcript?

Data reduction is the process of selecting and simplifying the data that you have collected.

During the data reduction process you need to decide what in the data is important, detect which patterns are most relevant to your evaluation questions, and identify what story the data tell.

In the picture, we can see how the data has been reduced, and we will only keep what it is on purple, what it is really relevant for our evaluation

This process often takes the form of “coding,” codes are labels that helps categorize the data.

For example, if you are asking participants what their favorite parts of the museum were, you may “code” the data using the names of the different exhibits.

12 of 20

Codebook

12

Code	Definition	Key Terms
Awareness, Knowledge, or Understanding	Measurable demonstration of assessment of or change in awareness, knowledge, understanding of a particular scientific topic, concept, phenomena, theory, or careers central to the project.	Awareness, Knowledge, Understanding, Educate
Engagement or Interest	Measurable demonstration of assessment of or change in engagement/interest in a particular scientific topic, concept, phenomena, theory, or careers central to the project.	Engagement, engage Interest connecting, reconnect Foster
Attitude	Measurable demonstration of assessment or change in attitude toward a particular scientific topic, concept, phenomena, theory, or careers central to the project or one’s capabilities relative to these areas.	Attitude Appreciation Curiosity Love respect,

“Codes” are just a fancy term for a flag that you apply to the data. They help you to go back and identify places in which you saw something that will help you to answer your evaluation questions.

For example, if you want to know if your participants had an increase in knowledge, and you interviewed them, you can “code” or flag that part of the interview when they talk about the content that they learned in the program.

Or maybe you are looking for engagement by observing video of an exhibit. You may code instances in a video by noting the time stamp in which you see visitors read and interact with the exhibit.

It is useful to create a “codebook” when doing this type of analysis. A codebook will help you to organize the information such as the codes, your definition for those codes, and perhaps key terms that you are looking for in the transcript.

13 of 20

Data Displays

Word Clouds

Venn Diagram

Quotes

13

“A museum is a place where one should lose one’s head”

Renzo Piano

14 of 20

Wrap up/Highlights

Remember this is the FUN part.
Simple is good and you can always partnering with someone with more experience.
Graphs, tables, and statistical techniques.
Data reduction, and data display.

14

15 of 20

Conclusion

15

16 of 20

For further information

Phillips, T. B., Ferguson, M., Minarchek, M., Porticella, N., and Bonney, R. (2014). User’s Guide for Evaluating Learning Outcomes in Citizen Science. NY: Cornell Lab of Ornithology.

Diamond, J., Luke, J. J., & Uttal, D. H. (2009). Practical evaluation guide: Tool for museums and other informal educational settings. Lanham, Md: AltaMira Press.

16

17 of 20

In the next module you will be able to…

17

Determine what evaluation results say about your program and�
Understand how to make evidence-based changes to your program.

18 of 20

Stay connected

18

Visit us: https://research.ced.ncsu.edu/impactnc/

Dr. K.C. Busch

Assistant Professor of STEM Education

North Carolina State University | College of Education

kbusch@ncsu.edu

19 of 20

Credits

K.C. Busch

Lynn Chesnut

Regina Ayala Chávez

Aimee Fraulo

Kathryn T. Stevenson

Katy May

Lincoln Larson

Madeline Hinckle

19

20 of 20

20