Psychology Week 4 Assignment: Correlation Application and Interpretation

Correlation Application and Interpretation

Week 4 | Data Analysis and Application

Student: [Your name]

Course: [Course number]

Date: November 2026

Analysis method: Python (SciPy), using the 105 records transcribed from the provided grades data PDF.

Section 1: The Data Analysis Plan

The four quantitative variables are Quiz1 (number of correct answers on Quiz 1), GPA (previous grade point average), Total (total course points), and Final (number of correct answers on the final exam). They are treated as continuous/scale variables for the Pearson correlation analysis, although the test scores are recorded as integer counts.

Research question 1: Is there a statistically significant linear relationship between Total and Final? H₀: ρ(Total, Final) = 0. H₁: ρ(Total, Final) ≠ 0.

Research question 2: Is there a statistically significant linear relationship between GPA and Quiz1? H₀: ρ(GPA, Quiz1) = 0. H₁: ρ(GPA, Quiz1) ≠ 0. Both tests use a two-tailed significance level of α = .05.

Section 2: Testing Assumptions

Table 1. Descriptive statistics and normality indicators (N = 105).

Variable	N	Mean	SD	Skewness	Excess kurtosis
Quiz1	105	7.467	2.481	-0.851	0.162
GPA	105	2.862	0.713	-0.220	-0.688
Total	105	100.086	13.427	-0.757	1.146
Final	105	61.838	7.635	-0.341	-0.277

Skewness indicates distribution asymmetry; excess kurtosis indicates tail weight relative to a normal distribution. Values near zero suggest approximate normality, but skewness and kurtosis alone do not prove normality. Quiz1 has a ceiling at 10 and some low scores, so its distribution deserves particular caution. The variables do not all appear strictly normally distributed; normality should not be considered fully established on these statistics alone. Pearson correlation also assumes approximately linear relationships, independent observations, and no influential outliers; these require scatterplot inspection and knowledge of data collection.

Section 3: Results and Interpretation

Table 2. Pearson intercorrelation matrix (two-tailed tests; N = 105).

	Quiz1	GPA	Total	Final
Quiz1	1.000	0.152	0.797	0.499
GPA	0.152	1.000	0.318	0.379
Total	0.797	0.318	1.000	0.875
Final	0.499	0.379	0.875	1.000

Total and Final: Pearson’s r(103) = 0.875, p < .001. The relationship is strong and positive. At α = .05, reject the null hypothesis.

GPA and Quiz1: Pearson’s r(103) = 0.152, p = 0.121. The relationship is weak and positive. At α = .05, fail to reject the null hypothesis.

Section 4: Statistical Conclusions

The analysis evaluates whether overall course points are associated with final examination performance and whether prior GPA is associated with Quiz 1 performance. The observed Pearson correlations quantify the direction and magnitude of linear association; statistical significance is assessed with two-tailed tests. These results do not establish causation.

Limitations include the use of a single course sample, possible ceiling effects on Quiz1, potential nonnormality, and the absence of a formal assessment of influential observations and linearity. Total points may include final exam points, creating a part–whole overlap that can inflate the Total–Final correlation. Other factors such as attendance, prior preparation, instructional differences, and study habits could influence scores. Generalization beyond these students should be cautious.

Section 5: Application

In industrial-organizational psychology and human resources, a useful application would be examining the correlation between employee job satisfaction scores and turnover-intention scores. Studying their relationship could help HR professionals identify workforce concerns and develop targeted retention strategies. A correlation could guide further investigation, but would not show that satisfaction alone causes employees to stay or leave.

References and Data Source

Capella University. (n.d.). Grades [Course-provided instructional data set].

Method note: Descriptive statistics use sample SD (n − 1) and bias-corrected skewness and excess kurtosis. Pearson correlations and two-tailed p values were computed with SciPy. This is a Python-based reconstruction of the required tables, not an export from JASP. Confirm that alternative software output is permitted and transfer content to the official DAA template if required.