Hello, if you have any need, please feel free to consult us, this is my wechat: wx91due
Psych 248 – Spring 2025
Computer Assignment 1
Descriptive Statistics
The goal of this computer assignment is for you to conduct descriptive statistical analyses and write up a report, like a real experiment! Please follow the instructions below to complete the assignment.
Instructions:
1. Enter data, conduct analyses in JASP and Excel, and save your output for your own reference.
2. In a separate word document, write up your results into a report to describe your results.
a. Copy and paste only the requested results into this report.
b. Write up your answers in paragraph format - just as you would a formal research report. DO NOT SUBMIT your answers in this document.
c. Please save your final written report in PDF or .docx format. Please do not submit a Google Doc.
Notes:
• You do not need to submit any Excel, .csv, or JASP files.
• You may work with others on transforming data and running analyses, but each student is expected to work independently on written responses.
• Do your best! Do not plagiarize! I have a zero-tolerance policy for plagiarism, and you will receive a zero.
• Please follow the directions provided here to create the results requested. Save JASP data and output files for later reference when you are writing up your results.
• Try to complete all data analyses during lab, then complete the written report later.
DUE DATE: FEBRUARY 24th at 11:59PM in the Microsoft Teams drop box.
• Late submissions are accepted with 1 point penalty per day late.
• I will not accept late assignments past March 3rd at 11:59pm.
• Do ask questions, save your files often, copy/paste your results and organize your written report in two parts.
• This assignment is worth 15 points.
You will submit one written report (for Parts 1 & 2) with only the requested graphs/tables.
Part 1: Data entry and descriptive analyses - A study of college students’ studying habits, GPAs & absences
You are an educational psychologist interested in the association between GPA, the number of hours students spent studying per week, and the number of absences from school. You collected data from educational records detailed below. Group 1 studied 5 hours or less each week and Group 2 studied more than 5 hours per week.
|
Group |
GPA |
Absences |
Group |
GPA |
Absences |
|
1 |
3.83 |
1 |
2 |
3.33 |
2 |
|
1 |
2.10 |
9 |
2 |
3.91 |
3 |
|
1 |
2.45 |
7 |
2 |
2.70 |
1 |
|
1 |
3.00 |
3 |
2 |
2.35 |
8 |
|
1 |
2.22 |
18 |
2 |
4.00 |
1 |
|
1 |
3.54 |
4 |
2 |
2.67 |
4 |
|
1 |
2.71 |
5 |
2 |
2.78 |
10 |
|
1 |
3.23 |
1 |
2 |
3.30 |
5 |
|
1 |
2.42 |
3 |
2 |
1.89 |
9 |
|
1 |
2.02 |
9 |
2 |
2.96 |
5 |
|
1 |
1.13 |
12 |
2 |
3.55 |
0 |
|
1 |
2.14 |
10 |
2 |
3.22 |
9 |
Instructions for Part 1 : Start a new doc for results – paste ONLY REQUESTED tables/graphs
1. Enter the data into an Excel file (hint: by row) and save as a .csv file. Review the data for any errors. Open the file in JASP.
2. Add labels for the independent variable (hint: you will have to change it to nominal before you can add labels).
3. Run descriptive statistics for each dependent variable (hint: only your DVs should be in the variables box).
4. Split the file based on your grouping variable (IV). Create a boxplot and histogram with density curve. Label outliers.
5. Check for normality (hint: add skew & kurtosis to stats table) and check distribution plots. a. Remember – for skewness:
i. If the skewness is lower than -1 (negatively skewed) or greater than 1 (positively skewed), the data is highly skewed.
b. For kurtosis:
i. >1 will indicate positive excess kurtosis (sharp/skinny peak).
ii. < -1 will indicate negative excess kurtosis (flatter peak).
iii. The greater the value of kurtosis, the higher the peak.
6. Create z-scores for both dependent variables.
a. Use + to add a Computed Column
b. Click + to Create Computed Column → Name the new variable → z(“Dependent Variable”), → click to Create.
c. To program in JASP, you must drag in functions - not type them in:
d. Drag in the zScores(y) function on the right-side column, then drag the dependent variable into the parentheses (y).
e. Click Compute Column, then click X to close out of the programing box f. Repeat for the second DV.
g. If you have problems using JASP to compute z-scores, use Excel to create z-scores.
The formula is similar. Start in cell E type in: = (C2 - M)/SD then copy/paste it down the column.
You should now have a data file with 6 variables (columns) - including 2 new z-score versions for GPA & Absences.
7. Look at descriptive statistics again including the Z-scores. (Notice that the mean of both z vars is close to 0, the SD = 1, shape is not changed.)
a. Run descriptives again with skewness and kurtosis for all 4 scale variables.
b. Copy the descriptive table to your word document.
8. Remove skewness and kurtosis, then run descriptives again for GPA and Absences, split by Group.
9. Under plots add Boxplots (Boxplot element) – should show GPA, Absences by Group
a. Add a title: “Your name - Descriptive Stats “by Study Group”, then copy this stat table & boxplots to your word document.
WRITTEN Qs for Part 1: Please use an 11-point font (Arial, Aptos, or Calibri) – make it organized and easy to read.
Remember to write up your answers in sentence format (do not write in bullet point format).
1. Predictions (1 point): Describe what this study is about and summarize your predictions (3-5 sentences).
a. Do you think the average GPA of those studying more than 5 hours will be higher or lower than those studying more than 5 hours?
b. How do you think the number of days absent of those studying 5+ hours will compare with those studying 5 or less hours?
c. What kind of relationship would you expect between grades and absences and why?
2. Overall Results (2 points): Describe the overall results including center, spread and
distribution shape (hint: does the distribution appear normal, skewed, describe the peak) (4-6 sentences).
a. What is the evidence for the approximately normal distribution of GPA and Absences (hint: what stats do we look at)?
b. Identify any outlier scores.
c. What are the IDs of the z-score values that are more than 2 SDs from the mean?
d. Which cases were the highest and lowest values for GPA & absences?
3. Results by Group (2 points): Compare the GPA and Absences of the two groups of
students (focus on central tendency & spread) (3-5 sentences).
a. Use the boxplots to discuss the amount of overlap between groups. How well do the data fit your predictions? Do you have any suggestions for future research on this topic?
PART 2: Analysis of a large dataset - A New York State Health Survey
A random sample of New Yorkers in large NY cities were surveyed about their health.
Instructions for Part 2:
1. Open the ‘ Health Data.csv’ file into JASP. Look at all of the variables, then drag all scale and ordinal variables into the Variables box. Run descriptive statistics, including skewness and kurtosis to check for normality.
a. Create frequency tables for the ordinal variables. Save this JASP data file.
2. Make some predictions. Which demographic variables (age, gender, marital status) do you think will be most strongly related to health behaviors (BMI, blood pressure, weight)?
3. Create a histogram with the density curve for BMI and systolic BP. Do these DVs appear normally distributed? Use statistics to backup your conclusions about normality.
4. Examine the effects of smoking, exercise, gender, and marital status on BMI and BP. a. Select BMI and BP as DV and run split file + descriptives for each of the above IVs.
i. Be sure to include variance, skewness, and kurtosis in your descriptive stats.
b. Use the default Descriptive Stats table and Plots (default boxplot). Copy these results to your report – Part 2 – BMI.
c. Which of these predictors seem to have effects on BMI and BP and which do not? Any surprises based on your predictions?
5. Do health outcomes differ among cities in New York State? Select systolic BP and BMI as
DVs. Run descriptives split on city.
a. Use the default Descriptive Stats table and Plots (default boxplot). Copy these results to your report – Part 2 – BP.
b. Which of health outcomes vary greatly by city? Which do not?
WRITTEN Qs for Part 2: Please use an 11-point font (Arial, Aptos, or Calibri) and make it organized and easy to read.
Remember to write up your answers in paragraph format!
1. Briefly describe the respondents in the study and the variables of interest in this
analysis (1 points). (2-5 sentences)
a. Summarize patterns found in the demographic variables (hint: look at frequencies - how many were married, how many vaped/used cigarettes, etc.) Are the 3 health DVs approximately normally distributed – BP, BMI, and weight?
2. Describe your predictions – which of the demographics (age, gender, marital status) did you expect to be most strongly related to health (systolic blood pressure, weight, BMI)? (2 points) (3-6 sentences)
3. Compare the boxplots looking at predictors of BP and BMI (3 points). (3-6 sentences)
a. Which factors seem to have no effect on BP and which ones have some effect?
b. Which factors seem to have no effect on BMI and which ones have some effect?
c. Describe any patterns you see. Do any of the results surprise you or differ from your predictions? Do you think any small n’s could affect some of these results?
4. Form some conclusions using the analyses and questions in #5 on pg. 4 about the
difference in BP and BMI between NY cities (2 points). (2-4 sentences) {Keep in mind these are survey data, not experimental data.} What do the data tell us about differences between NY cities?
5. Can you think of some other interesting comparisons that could be made using any of
the variables in this data set? (+1 Extra Credit)
a. Come up with at least one new question and then suggest what kind of statistical approach could be used to explore the question or test your prediction.
b. Keep in mind this is observational data, not experimental (we cannot test causation!)