Psi Beta National
Research Study
How to prepare the data file. How to analyze your data.
The Psi Beta research team at Irvine
Valley College
• Tyler Wong - Research team leader
• Ruth Castillo
• Sofia Parrish
• Shireen Mohamdjawad
• Brittany Kester
Psi Beta’s National Research
Committee
• Heather Schoenherr (Committee chair &
Rocky Mt. Vice President)
• Liza Veliz (Southwest Vice President &
National President-Elect)
• Elizabeth Arnot-Hill (Midwest Vice
President)
• Jerry Rudmann (Psi Beta’s Executive
Director)
ACKNOWLEDGEMENTS
Access and Use of Psi Beta’s National Research
Study Data Files
Who can use the data?
All psychology teachers and their students
(regardless of Psi Beta membership) are
encouraged to use the data, especially chapters
that helped to recruit participants.
If the data are used for a poster presentation or
research paper, please acknowledge Psi Beta
National as the data source.
3
5
6
Study Overview…
• The study involved a questionnaire administered online.
• 121 items
• 9 scales – chosen for potential to measure changes
caused by how colleges and students adjusted to the
pandemic safety measures.
• We used Google Forms to create the questionnaire.
• Easy and free. If you don’t know how to use Google
Forms to make a survey, look for YouTube videos.
• The form’s data is put into a Google spreadsheet.
• We saved the Google spreadsheet as an Excel file. That
file – we call the “raw data” file can be downloaded from
the Psi Beta website.
• A PDF version of the entire questionnaire can be
downloaded from Psi Beta’s website.
• You’ll need a copy. It will help you keep track of what
each item is. Also, it specifies “reverse coded” variables
that you’ll need to recode. (I’ll explain that later.)
• N = 1,530 responses (but not all responses are useful)
Let’s look at the
PDF copy of the
online
questionnaire used
for this study
You’ll be able to download this from the Psi
Beta website.
8
The Scales and Demographics
Scales
1.
Self-efficacy for initiating conversations
2.
Interest in learning interpersonal communication
skills
3.
Efficacy for facilitating balanced two-way
conversations
4.
Interpersonal competence during COVID-19
5.
College connectedness
6.
Ten-item Big 5 Inventory
7.
Shyness
8.
Social media usage aims
9.
Social media use questionnaire
10.
Most preferred social media platform (a single item,
not a scale)
11.
Estimated time spent on favorite social media
platform per day (a single item, not a scale)
Demographic Variables
• College
• Sex
• Age
• # semesters in college
• FT / PT student
• Racial/ethnic background
• Mother attend college
• Father attend college
• Any sibling attend college
• Intended academic major
• Honor society member
9
Software needed to prepare the data file for analysis
• Excel
• JASP or SPSS. JASP is free. At the end today, I’ll show you how to get a free
copy of JASP. Next week, I’ll show you how to use JASP to analyze the data file
you have prepared.
Files you’ll need to download
• “Raw” data file (Excel)
• PDF of the research questionnaire (what the participants responded to when
they participated). Each question on the questionnaire has a suggested
variable name in red. Some items will need to be “reverse coded” – those
items have a green
R
next to them.
• Code book – a guide to renaming each item
• The Ten-Item Personality Inventory (TIPI)
Accessing and Preparing the Raw Data
10
Steps for Preparing & Using the National Research Study Datafile
“Raw” Data
• It is a spreadsheet
composed of cases and
variables
• It has some unneeded
columns (i.e., variables)
• It has some incomplete
cases
•
It may contain “suspicious”
cases
• It has some items with
words (e.g., “Agree”
“Disagree”) that need to be
converted to numbers
• Some items need to be
recoded (“reverse coding”)
Clean & Prepare Data
-
Use
Excel
• Column and case removal
• Columns with useless
data
• Cases with too much
missing information
• Rename variables
• Refer to the PDF
questionnaire and the
codebook as you rename
the variables on your raw
data spreadsheet
• Recode string variables to
numerical variables
• Reverse code Likert items
• Delete suspicious or “test”
cases
Conduct Data Analyses
• Export Excel file to CSV
format
• In JASP – make sure all
variables are numerical
• In JASP or SPSS - Compute
scale and subscale scores
• Describe participants
(compute Descriptive stats)
• Run statistical tests to
evaluate your hypotheses
11
Steps for Preparing & Using the National Research Study Datafile
“Raw” Data
• It is a spreadsheet
composed of cases and
variables
• It has some unneeded
columns (i.e., variables)
• It has some incomplete
cases
•
It may contain “suspicious”
cases
• It has some items with
words (e.g., “Agree”
“Disagree”) that need to be
converted to numbers
• Some items need to be
recoded (“reverse coding”)
Clean & Prepare Data
-
Use
Excel
• Column and case removal
• Columns with useless
data
• Cases with too much
missing information
• Rename variables
• Refer to the PDF
questionnaire and the
codebook as you rename
the variables on your raw
data spreadsheet
• Recode string variables to
numerical variables
• Reverse code Likert items
• Delete suspicious or “test”
cases
Conduct Data Analyses
• Export Excel file to CSV
format
• In JASP – make sure all
variables are numerical
• In JASP or SPSS - Compute
scale and subscale scores
• Describe participants
(compute Descriptive stats)
• Run statistical tests to
evaluate your hypotheses
12
Raw Data Spreadsheet (Excel)
13
Raw Data Spreadsheet (Excel)
14
A. Rename each variable to match the codebook
B. Remove cases with too much missing data - Use eyeball or
use Excel’s “Countblank” formula to count number of blank cells
per participant
C. Recode string variables to numerical variables - Use Excel’s
“find and replace” feature to change words to numbers
D. Recode reverse coded variables
Here’s what we will do, in this order….
15
Rename your variables…
16
Rename your variables…
17
Rename Items
18
Remove bad cases, for example…
- Cases with too much missing data (i.e., blank cells)
- Cases from unknown college
- Cases showing a “response set”
19
• Use Excel’s “CountBlank” formula
Finding Cases With Missing Data
After finding cases with blanks, delete cases with too much missing data
20
Sequence
1. Highlight all cells
2. Click on “Data” on the top
horizontal menu
3. Select the sort icon to bring up a
dialog box
4. In the dialog box Sort by option,
choose BLANK_CELLS
5. In the dialog box Order option,
choose Largest to Smallest option,
then click on OK a the bottom of
the dialog box
This will reorder all cases and put the
number of cases having the most
missing cells at the top of the
spreadsheet
Use Excel’s sort feature to find the cells with too much missing information.
21
22
Now, let’s remove all words from your numerical variables
Use Excel’s Find and Replace dialog box
23
Note, for example, that items T1 through T10 have words than need to be recoded
into number
24
Step-by-step for replacing words with numbers
25
Step-by-step for replacing words with numbers
26
27
Items T1 through T10 are now recoded from words to numerals.
28
Other spreadsheet items to consider recoding from words to numbers:
1. FT_PT
Full-time (enrolled in 12 or more units this semester) = 2
Part-time (enrolled in less than 12 units this semester) = 1
2. Honor_Society
Yes = 2
No = 1
3. Mom_College, Dad_College, Sib_College
Yes = 2
No = 1
Not sure = 1
Not applicable - 1
29
Age item. In your spreadsheet, scan through the Age column
- Change years to ages
- Remove silly ages
- Sort the spreadsheet based upon the Age columns and remove 17 &under
participants (there are around 70 below age participants)
30
Now, let’s reverse code your variables
31
Reverse Coding – Step by Step
32
Reverse Coding – Step by Step
33
Reverse Coded Variable - example
Q56 – I am usually a person who initiates a conversation.
o 1 (Very Uncharacteristic)
o 2
o 3
o 4
o 5 (Very Characteristic)
This item needs to be reverse coded for the Shyness Scale.
34
Reverse Coding the Shyness Scale items
o 1 (Very Uncharacteristic)
o 2
o 3
o 4
o 5 (Very Characteristic)
o 1 = 5
o 2 = 4
o 3 = 3
o 4 = 2
o 5 = 1
Original
After Reverse Coding
35
Reverse Coding for the T-Scale items
o 1 Disagree strongly
o 2 Disagree moderately
o 3 Disagree a little
o 4 Neither agree nor
disagree
o 5 Agree a little
o 6 Agree moderately
o 7 Agree strongly
o 1 = 7
o 2 = 6
o 3 = 5
o 4 = 4
o 5 = 3
o 6 = 2
o 7 = 1
Original
After Reverse Coding
36
Here are the items that need to be reverse coded in this year’s data file:
S4, S7, S10,
S13, S16,S18
T2, T4, T6,
T8, T10
37
Reverse Coding – A Step-By-Step Example to reverse-code item T2
• Find item T2 on the data Excel sheet.
• Add a blank column to the very right of the T2 column.
Right click on T3 and insert an entire column. This adds
a blank column to the right of the T2 column.
• Label the new blank column T2R.
• Place the cursor in the first cell of T2R’s column (row 2).
• Enter this formula =8-AO2 and press the enter key. You should see
reverse code of the original number (i.e., 2 is now 4). Note that
when you drag down to the rows below, Excel automatically
changes your formula (e.g., “=8-AO2” changes to “=8-AO3” in the
next cell down, and so on).
• Repeat this procedure for all variables in the codebook that need
recoding. If you see an R in the first column of the codebook, the
adjacent variable needs to be recorded.
NOTE: Do NOT delete the T2 column. Leave it be
.
38
You’ll need to adjust the recode formula for items on the Shyness Scale.
The T-Scale had 7 possible responses. So,
the recode formula for item T2R is
=8-AO2
The recode formula for T4R will be
=8-AR2
And so on for T6, T8, and T10
Each Shyness Scale item has just 5
possible responses.
The formula for item SR4 will be =6-BC2
and so on for the remaining Shyness
items that must be reverse coded.
As you may have noticed, the recode
formula always uses a number 1-digit
higher than the number of alternatives
for the survey question. Pretty cool!
39
When your file is ready, save it, then save it as a CSV file. JASP works with CSV files.
Your file will be ready for next week’s workshop after you have
• Renamed your variables on the top row of the spreadsheet
• Removed cases with too much missing data
• Removed cases with participants who are younger than 18 years
• Recoded string variables to numbers
• Recoded reverse coded items
40
Using JASP
• Save your cleaned up Excel file as a
CSV file
• Download JASP at jasp-stats.org
• Next week I’ll show you how
• Import the CSV file into JASP
• Use JASP to compute total scores for
the various scales:
• Shyness
• Connectedness
• Extraversion
• All other scales
• JASP provides statistical procedures
you can use to analyze your data
(not hard to do!)
41
Your assignment: Try to do this before next Friday at 11 am PST
Become Familiar
with Google Forms &
Spreadsheets
Download & prepare
the “raw data” file
from the National
Research Study
Download and install
J A S P. W a t c h a f e w
You Tube help videos
on JASP
42
JASP’s Opening Screen
43
Knowing this information and having these skills = GREAT VALUE!
CREATE ONLINE
QUESTIONNAIRES &
SURVEYS
PREPARE THE RAW DATA ANALYZE THE DATA COMMUNICATE RESEARCH
FINDINGS TO NON-
SCIENTISTS
44
Thank you!
See you next Friday…Jerry Rudmann
jerryrudmann@gmail.com
