A small survey, start to finish: from frequencies to regression

Raw data walkthrough_survey.csv · 60 rows

Synthetic data — the 60-row survey built by scripts/walkthrough/make_sample.py, the same sample the guide articles and the walkthrough use. Every table and APA sentence is a measured result: backend/tests/test_en_sample_walkthrough_en2.py runs the engines in English to produce them.

The study, step by step — what was actually produced

1. ① Frequencies — who answered

   Start by counting: how many people in each gender and each department, and whether any cell is missing. Skip this and every later test reports a result without you ever knowing how many cases it ran on.

   A frequency analysis was conducted on gender and department (N = 60). For gender (2 categories), "F, M" each appeared equally often (30, 50.0% of valid cases); valid n = 60, missing = 0. For department (3 categories), the most frequent category was "Research" (24, 40.0% of valid cases); valid n = 60, missing = 0. No missing data were present.

2. ② Descriptives — the shape of the score

   Mean, standard deviation, median, quartiles, range, skewness and kurtosis for the score column. A distribution piled up on one side changes which test belongs here.

   Descriptive statistics were computed for score (N = 60). For score, M = 3.75, SD = 0.77, Mdn = 3.96, range = 1.73–5.00. Skewness = -0.57 and kurtosis = -0.05, within normal range. No missing data were present.

3. ③ Independent-samples t-test — does it differ by gender

   Compare two group means. Levene’s test checks the equal-variance assumption alongside it, and the result carries an effect size (Cohen’s d) and a confidence interval for the difference.

   Independent-samples t-tests compared score across levels of gender (N = 60). For score, F (M = 4.11, SD = 0.66) scored significantly higher than M (M = 3.39, SD = 0.71), Student's t(58) = 4.06, p < .001, d = 1.05, 95% CI [0.36, 1.07]. Levene's test indicated equal variances.

4. ④ One-way ANOVA — does it differ by department

   With three groups you do not run three t-tests; you run one. When the effect is significant, the post-hoc comparison (Tukey HSD) names the pair — that table is printed here too.

   One-way ANOVAs examined score across levels of department (N = 60). For score, there was a non-significant effect, F(2, 57) = 1.77, p = .180, η² = .06. Levene's test indicated equal variances (p = .886). Group statistics: Research (M = 3.96, SD = 0.71), Sales (M = 3.51, SD = 0.80), Support (M = 3.70, SD = 0.79).

5. ⑤ Correlation — what moves together

   Three variables make three pairs, so the result is a matrix rather than a sentence: each cell carries r with its p.

   A Pearson correlation analysis was conducted (N = 60). score and age were significantly correlated in a positive direction, r = .26, 95% CI [.01, .49], p = .041. score and satisfaction were significantly correlated in a positive direction, r = .87, 95% CI [.79, .92], p < .001.

6. ⑥ Regression — what predicts the score

   Predict score from age and satisfaction. Gender is text, so the engine leaves it out of the regression — and says so in a footnote instead of dropping it quietly.

   A multiple linear regression was conducted to predict score from age, satisfaction. The overall model was significant (R² = .77, adjusted R² = .76, F(2, 57) = 93.64, p < .001). The predictors accounted for 76.7% of the variance in score. age was not a significant predictor (β = .11, p = .083). satisfaction significantly predicted score (B = 0.72, SE = 0.06, β = .85, t(57) = 13.05, p < .001). Multicollinearity was assessed using the variance inflation factor; the maximum VIF was 8.28 (age), indicating no serious multicollinearity. · Excluded variables: gender — not numeric, so they were left out.

Key results

Representative quotes

Every quote is a sentence that exists in the raw data (excerpt-existence check passed).

Paper-ready tables