10Min Research - 38 (P2) - How to Write the Discussion Section/Chapter - Part 2

A strong discussion section has to do more than restate results: it must explain why findings are significant, why they aren’t, and what the outcomes mean for theory and practice. When results come out insignificant—or even contrary to expectations—the write-up should still provide a clear, evidence-based rationale. That means summarizing the specific hypothesis and then laying out plausible explanations for the lack of statistical significance, such as insufficient sample size, random variation, confounding variables, measurement error, or an inappropriate statistical test. The goal is to show that the outcome is interpretable, not dismissible, and to clarify how the result affects what is currently understood about the research question.

The discussion also needs to handle “unexpected” directions carefully. A negative relationship can be statistically significant or not, but either way it should be treated as a meaningful contradiction if the study expected a positive effect. In the servant leadership example, the expected positive link to life satisfaction did not appear; instead, the relationship turned negative. The proposed reasoning is grounded in how servant leadership may shape behavior at work and spill over into personal life. One explanation is that person-oriented servant leadership can strengthen safe, strong workplace relationships—making employees enjoy their jobs—while simultaneously reducing their ability to enjoy other life domains. Another explanation is that servant leadership may encourage higher effort and career focus, which could increase exhaustion and push employees toward workaholism, ultimately lowering life satisfaction.

Theory can also be used to make non-significant results feel less like a dead end and more like a test of boundary conditions. In an authentic leadership and employee creativity example, psychological safety is used as a theoretical lens to explain why the expected relationship did not emerge. The argument is that authentic leadership may only translate into creativity when psychological safety climate is high—yet the study did not measure or control for that climate. The sample composition also matters: if employees came from many different teams, the study may not have represented the contexts where authentic leadership is most effective. These points naturally lead to future research recommendations: use larger, more representative samples and include relevant moderators (like psychological safety climate) and additional variables that could clarify when effects should appear.

Beyond interpretation, the discussion must connect findings to implications and limitations. Practical implications should translate results into actionable guidance for practitioners, while theoretical implications should specify what the study adds to existing knowledge. Limitations should be stated transparently—such as small sample size or limited generalizability—and tied to how they may have influenced results. Finally, future research directions should propose concrete next steps, including new mediators, moderators, antecedents, or outcomes that could better capture the mechanisms behind the relationships under study.

The transcript closes with structural guidance for writing: begin with the purpose and objectives, discuss each hypothesis one at a time in relation to prior research (supporting, contradicting, or extending it), then provide an overall conclusion aligned with the study’s aims. Strong discussion sections end by restating contribution, summarizing whether objectives were met, acknowledging limitations, and outlining what research should do next.