What effect does having a larger sample size for some vehicles compared with others have on the validity of the reliability data?
Given an appropriate sample, the more data you have, the more statistical confidence you have in your information. A larger sample will always give more accurate information than a smaller sample (assuming, of course, that the data are valid and collected from an appropriate source). While we require a minimum of about 100 cars to publish reliability information, most models have larger samples than that, some being as large as several thousand. We present our data primarily to allow subscribers to compare the detailed reliability histories and overall reliability for different models. While models whose scores are based on more cars are reported with greater accuracy than those based on smaller sample sizes, the way we calculate our scores has been devised to allow valid comparisons for all samples we publish, regardless of the particular sample sizes of individual models.
Related Questions
- I want to conduct some multivariate analyses using the EMI-2 but with so many subscales my sample size is not big enough. How can I reduce the data?
- What effect does having a larger sample size for some vehicles compared with others have on the validity of the reliability data?
- Where can I get norms and psychometric data (reliability/validity) for the DASS?
