Research methods exam tips

Exam tips

• Clarify data type in answers

  When answering questions about data handling, explicitly state whether the data is primary or secondary and justify your choice based on the research context.

  This demonstrates a clear understanding of data types and shows the examiner that you can apply the concepts appropriately.

• Show your calculations clearly

  When calculating mean, median, or mode, write out each step (e.g., sum of values, division, ordering). This demonstrates procedural knowledge and allows partial credit if a final value is incorrect.

  Examiners can see your reasoning and will award marks for correct steps even if the final answer is off.

• Check Units and Rounding

  Always report standard deviations to two decimal places unless the specification says otherwise, and include the correct unit (e.g., points, minutes).

  Consistent rounding and units demonstrate careful calculation and attention to detail, which examiners value.

• State direction and strength clearly

  When interpreting a correlation, always mention whether it is positive or negative and whether it is weak, moderate, or strong.

  This demonstrates a full understanding of the coefficient and helps the examiner assess your interpretation accurately.

• Choosing the right graph

  Always state the type of data (categorical or continuous) before selecting a graph. Use a histogram for continuous distributions, a scattergram for relationships between two continuous variables, and a bar chart for categorical comparisons.

  Selecting the appropriate graph ensures that the data are presented clearly and that the key features of the distribution or relationship are highlighted.

• Check distribution shape before choosing tests

  Plot a histogram or boxplot and calculate skewness. If the data are skewed, consider data transformation or non‑parametric tests.

  Ensuring that the assumptions of parametric tests are met prevents incorrect conclusions and demonstrates careful data handling.

• Report r, direction, strength, and significance together

  When presenting a correlation, state the coefficient, whether it is positive or negative, classify its magnitude, and note whether it is statistically significant, including the p‑value if required.

  This comprehensive reporting demonstrates understanding of the concept and satisfies examiners’ expectations for clarity and accuracy.

• Choosing the right statistical test

  Always check the level of measurement of your data before selecting a test. Use non‑parametric tests for ordinal data and parametric tests only when interval or ratio assumptions are met.

  Matching the test to the data properties prevents invalid statistical conclusions and strengthens the credibility of your findings.

• Plan your coding scheme carefully

  Draft a comprehensive coding frame, pilot it on a small sample, refine it, and then calculate inter‑coder reliability before coding the full dataset.

  A well‑planned scheme reduces ambiguity, improves reliability, and saves time during the coding phase.

• Structure your answers clearly

  When answering questions about statistical testing, first state the null and alternative hypotheses, then describe the test used, report the p‑value, and finally interpret the result in terms of statistical significance. This logical flow demonstrates understanding and helps you earn full marks.

  Examiners look for a systematic approach that shows you know the purpose of each component of inferential testing and can link them to the conclusion.