Stats and Methods Urban Legend 2: Control Variables Improve Your Study
The use of control variables to purify statistical analyses is most often an invalid approach to solving the problem of poor methodology and design.
Category: research methods
Stats and Methods Urban Legend 1: Formative Measurement
There are two models of the relationships between constructs and measures: reflective and formative. And formative’s got some issues.
Does Personality Predict Vulnerability to Violence in Games? (VG Series Part 2/10)
Part 2 of my series examining research evidence for the value of video games. This time: a spherical model of personality for predicting video game violence.
Some Employers Ruin Surveys For the Rest of Us
Some employers survey their employees “anonymously” but act dishonestly. What do the rest of us do to convince employees of our truthfulness?
Your Genetics Incline You to Respond to Surveys
Genetics play a role in predicting if you’ll respond to surveys. Over 1000 twin pairs were contacted through the Minnesota Twin Registry, and it was found that 45% of the variance in survey response behavior could be explained by genetic differences.
The Lies That Data Tell
Without a clear rationale for doing so, statistical significance testing on sample-level statistics can mislead and confuse. Schmidt (2010) provides a clear explanation of how to avoid this problem through psychometric meta-analysis.