Exam Anxiety: Descriptive Plots and Correlations

Description:


This fictional data set, "exam_anxiety", contains data from 103 participants, investigating the effects of exam stress on exam performance. Anxiety was measured before an exam, and then the percentage mark of each student on the exam was used to assess the exam performance. This is the data set used in Chapter 5 and 7 of the Discover Statistics using JASP textbook.


Variables:


  1. id: participant number
  2. Time spent revising: time
  3. Exam performance: percentage scored on the exam
  4. Exam anxiety: score on a questionnaire out of 100
  5. Gender: participant gender


This example JASP file demonstrates the use of descriptive plots.


Reference:


Field, A., van Doorn, J., & Wagenmakers, E.-J. (2025). Discovering statistics using JASP. London: Sage. [Fictional data set]


This data set comes from Field, A. P. (2025). Discovering Statistics Using JASP. London: Sage. The data set was constructed by Andy Field who therefore owns the copyright. Andy Field generously agreed that we can include the data set in the JASP data library. This data set is also publicly available on the website that accompanies Andy Field's book https://study.sagepub.com/field6e. Without Andy Field's explicit consent, this data set may not be distributed for commercial purposes, this data set may not be edited, and this data set may not be presented without acknowledging its source (i.e., the terms of a CC BY-NC-ND license see https://creativecommons.org/licenses/by-nc-nd/3.0/).

Descriptive Statistics: Simple Scatterplot

Descriptive Statistics
  Exam anxiety Exam performance
Valid 103 103
Missing 0 0
Mean 74.344 56.573
Std. Deviation 17.182 25.941
Minimum 0.056 2.000
Maximum 97.582 100.000

Scatter Plots

The output below shows the resulting scatterplot. The scatterplot and the descriptives indicate that the majority of students suffered

from high levels of anxiety (there are very few cases that had anxiety levels below 60, and the mean is 74). There also seems to be

a trend in the data, that higher levels of anxiety are associated with lower exam scores and low levels of anxiety are associated with high examination marks.

Exam anxiety - Exam performance

Association Matrix

Correlation
  Exam anxiety Exam performance
Exam anxiety 1.000 -0.441
Exam performance -0.441 1.000

Descriptive Statistics: Matrix Scatterplot

Descriptive Statistics
  Time spent revising Exam performance Exam anxiety
Valid 103 103 103
Missing 0 0 0
Mean 19.854 56.573 74.344
Std. Deviation 18.159 25.941 17.182
Minimum 0.000 2.000 0.056
Maximum 98.000 100.000 97.582

Correlation plot

From the matrix above, we can see that: time spent revising and exam anxiety are inversely related (upper right corner).

Q-Q Plots

The QQ Plots below show that the variables are not normally distributed, time spent revising and exam anxiety being clearly skewed.

Time spent revising

Exam performance

Exam anxiety

Descriptive Statistics: Split Lines Scatterplot

Descriptive Statistics
Exam anxiety
Exam performance
  Man Woman Man Woman
Valid 52 51 52 51
Missing 0 0 0 0
Mean 74.384 74.303 56.692 56.451
Std. Deviation 16.422 18.088 26.318 25.811
Minimum 10.000 0.056 2.000 5.000
Maximum 97.582 95.970 100.000 100.000

Scatter Plots

The figure below shows the resulting scatterplot. It is the same as in the previous plot, but there are separate lines for each of the genders (man vs woman). These show that the relationship between exam anxiety and exam performance was slightly stronger in men. However, the confidence intervals that JASP adds around the lines suggest that this difference might not be significant, since they overlap considerably (an important note is that the confidence intervals here exceed the score of 100 on the exam, which is outside the scale, thus they should be interpreted with caution).

Exam anxiety - Exam performance

Correlation

Pearson's Correlations
Variable   Exam performance Time spent revising Exam anxiety
1. Exam performance Pearson's r
p-value    
Lower 95% CI
Upper 95% CI
2. Time spent revising Pearson's r 0.397
p-value < .001  
Lower 95% CI 0.220
Upper 95% CI 0.548
3. Exam anxiety Pearson's r -0.441 -0.709
p-value < .001 < .001
Lower 95% CI -0.585 -0.794
Upper 95% CI -0.271 -0.598

Correlation plot

From the correlation plots above, we can see a strong negative (Pearson's) correlation, r = −0.441 between exam anxiety and exam performance and the significance value of this correlation p < 0.001 can also be seen in the table below.

Nonparametric Correlations

Correlation Table
Pearson
Spearman
Kendall
      r p Lower 95% CI Upper 95% CI rho p Lower 95% CI Upper 95% CI tau B p Lower 95% CI Upper 95% CI
Exam performance - Exam anxiety -0.441 < .001 -0.552 -0.304 -0.405 < .001 -0.551 -0.230 -0.285 < .001 -0.397 -0.162
Exam performance - Time spent revising 0.397 < .001 0.248 0.517 0.350 < .001 0.166 0.512 0.263 < .001 0.129 0.380
Exam anxiety - Time spent revising -0.709 < .001 -0.869 -0.424 -0.622 < .001 -0.773 -0.441 -0.489 < .001 -0.616 -0.345
Note.  Confidence intervals based on 1000 bootstrap replicates.

Partial Correlation

The table below shows the partial correlation of exam anxiety and exam performance adjusting for time spent revising. The partial correlation between exam performance and exam anxiety is −0.247. This value is considerably less than when we didn’t adjust for the effect of revision time (r = −0.441). Although this correlation is still statistically significant (p = .012) and the confidence interval [−0.431, −0.034] still doesn’t contain zero, the relationship is diminished. The value of R2 for the partial correlation is 0.061, which means that exam anxiety now shares only 6.1% of the variance in exam performance (compared to 19.4% when revision time was not factored in). Exam anxiety alone does explain some of the variation in exam scores, but there is a complex relationship between anxiety, revision and exam

performance that might otherwise have been ignored.

Pearson's Partial Correlations
      Pearson's r p Lower 95% CI Upper 95% CI
Exam performance - Exam anxiety -0.247 0.012 -0.431 -0.034
Note.  Conditioned on variables: Time spent revising.
Note.  Confidence intervals based on 1000 bootstrap replicates.