Stress Reduction Training and Puzzle-Solving Performance

Study Design and Methodology

This study examined the influence of stress reduction training on cognitive performance under stressful conditions, with particular attention to how baseline life stress levels moderate this effect. The research design employed a 2×2 factorial structure with two independent variables and one dependent variable measured at the interval level.

Participants were randomly assigned to either a control group (n=15) or a treatment group (n=15). Before the experiment began, all 30 participants completed a life stress test that classified them as either high or low on baseline life stress. The treatment group received one hour of daily training for one week, during which they learned various evidence-based techniques for decreasing stress levels. The control group received no training during this period.

Following the one-week interval, all participants completed a puzzle-solving task under stressful environmental conditions (including loud noises and bright lights). The dependent variable was the number of puzzle pieces completed within a five-minute time limit, measured out of a total of 10 pieces. Researchers expected three distinct outcomes: that the training group would differ from the control group, that high-stress and low-stress participants would differ from each other, and that trained participants with high baseline stress would perform differently from untrained participants with low baseline stress.

The independent variables in this study were (1) training condition (treatment versus control) and (2) baseline life stress level (high versus low). The quantitative data collected consisted of performance scores on the puzzle task, measured in terms of puzzle pieces completed within the time frame. Factorial ANOVA (analysis of variance) was selected as the appropriate statistical method because the design included two or more categorical independent variables and a single normally distributed interval-level dependent variable (Lane, 2007). This design allows testing of main effects for each independent variable as well as their interaction.

The research question guiding the study was: What influence does stress reduction training have on the performance of people living with different levels of life stress? Two corresponding hypotheses were formulated. The null hypothesis (H0) stated that stress reduction training does not influence puzzle-solving performance under stressful conditions. The alternative hypothesis (H1) stated that stress reduction training does influence such performance. A small sample size (N=30) was acknowledged as a limitation that would restrict the generalizability of findings beyond the study population.

Descriptive Statistics and Data Characteristics

Descriptive statistics were computed for all groups to characterize the raw data before inferential testing. Descriptive statistics provide measures of central tendency and variability, which describe the shape and spread of distributions for interval and ratio variables. Two measures of distribution shape were examined: skewness (the degree and direction of asymmetry from horizontal symmetry) and kurtosis (the relative peakedness of the distribution compared to a normal bell curve).

The kurtosis analysis revealed heavy tails in the performance distribution, indicating that participants' scores tended toward either very rapid or quite slow puzzle-solving behavior, with fewer scores in the middle range. However, it is important to note that measures of shape such as skewness and kurtosis are unreliable estimators when sample sizes are small, as they were in this study (SAS, 2006).

Mean performance scores for the treatment group were 6.20 pieces (SD = 1.304) for the high-stress subgroup and 7.60 pieces (SD = 1.075) for the low-stress subgroup. In contrast, mean performance scores for the control group were 2.50 pieces (SD = 0.972) for the high-stress subgroup and 2.80 pieces (SD = 0.837) for the low-stress subgroup. This pattern shows a substantially larger effect of training than of baseline stress level.

When examining the main effects independently, the mean performance score for the training variable alone was 2.65 pieces for the control group and 6.90 pieces for the treatment group. For the life stress variable alone, the mean performance score was 4.35 for the control condition and 5.20 for the treatment condition. The descriptive data suggested that training had a more pronounced effect on performance than baseline life stress did, and that minimal interaction between these two variables existed.

Factorial ANOVA Results and Analysis

A factorial ANOVA was conducted to test the statistical significance of the main effects and interaction effects. The overall model was statistically significant (F = 48.949, p = 0.00), indicating that the combination of training and baseline life stress explained a significant amount of variance in puzzle-solving performance.

The main effect of training was statistically significant (F = 54.682, p = 0.000), with a mean square of 120.4 (f = 109.85, p < 0.05). This finding provides strong evidence that stress reduction training substantially improves puzzle-solving performance under stressful conditions. In contrast, the main effect of baseline life stress was not statistically significant (F = 4.394, p = 0.046). The interaction between training and baseline life stress was also not statistically significant (F = 1.840, p = 0.187), with a mean square of 2.02 (f = 1.84, p < 0.05).

No post-hoc tests were performed for training or life stress because post-hoc comparisons are unnecessary when independent variables have fewer than three levels (UCLA, 2010). The results clearly indicate that the effect of training is the dominant factor in explaining performance differences, while baseline stress level and the combined effect of both variables did not reach statistical significance.