7.3 Lab Report Instructions

Prepare an R script and written report presenting the results of the analysis described in the analytic strategy.

7.3.1 R Script

Your R script should be reproducible and free of errors. It should contain syntax to do the following:

  • Import the data and convert the “condition” variable to a factor.
  • Calculate participant mean scores on the 4-item national identity measure.
  • Compute descriptive statistics, including M, SD, skew, and kurtosis of participants’ mean national identity. Do this separately for each level of the experimental manipulation. Use by() and describe() to do this in a single step.
  • Conduct Shapiro–Wilk tests of normality for the distributions of national identity in each of the two conditions.
  • Compute a Welch’s t test as described in the analytic strategy.
  • Calculate the corresponding Cohen’s d effect size for the t test.

7.3.2 Results

The Results section should include the following:

  • Total participants in each experimental condition (note that the actual number of participants might differ from the planned number of participants in the study preregistration).
  • M and SD of national identity for each experimental group.
  • Results of the null-hypothesis significance test and corresponding standardized effect size with its associated confidence interval.
  • Description of efforts to diagnose problems with statistical assumptions and data distributions.
  • A figure visualizing mean national identity in each of the two groups and error associated with that measurement. The figure should be referenced in the text of the results section.
    • For example, boxplot, points with error bars, barchart with error bars.
    • If you use error bars, indicate what they represent in the note to the figure.

7.3.3 Discussion Questions

Provide short answer responses to the following:

  • Provide a statement of support or nonsupport for your primary hypothesis.
  • Provide an interpretation of the results. Consider both the theory under investigation and the design of the study. Other than the proposed theoretical mechanism, what design or statistical issues could bias the results?
  • Canadian adults are the implied population of interest in this study. Given the sampling design, how well does this sample represent that population? How does that affect the interpretation of these results?
  • What is an implication of this research for future research, programs, or policy?