Collected from multiple websites (wikipedia.org, tv.com and television show websites) this data set consists of observations of the ranking of contestants in live shows for several pop Idol series, from eight countries. The data were collected from the following countries during the specified years:

• Australia (Australian Idol: 2003, 2004, 2006, 2007; X-factor: 2005)
• Brazil (Idolos Brazil: 2007)
• Canada (Canadian Idol: 2003, 2004, 2005, 2006, 2007)
• Germany (Deutschland sucht den Superstar: 2003, 2004, 2006, 2007)
• India (Indian Idol: 2006, 2007)
• Netherlands (Idols: 2005; X-factor: 2006)
• UK (X-factor: 2004, 2005, 2006, 2007)
• USA (American Idol: 2002, 2003, 2004, 2005, 2006, 2007)

The effects of ordering on the evaluation of individual performance were studied, as the shows share a generic format. A comparison between the empirical probability to be "safe" during one show and the theoretical probability to be "safe" was conducted, to assess the existence of bias in contestants' performance evaluations. The publication arising from an analysis of this data presents new evidence about the systematic biases in sequential evaluation of performances and the psychological phenomena at the origin of these biases.

Analysis of the data was performed using Stata and SPSS. Identification systems in the data: variable to code "bottom 2."