PsychologyOrbit
Sign in Join

#4 ·

Spot the flaw #4: How common is smartphone addiction?

A past puzzle, with the answer. Try today's puzzle.

What is the flaw in this study?

Selection bias in the sample. People who visit a digital-detox forum are, almost by definition, worried about their phone use. Recruiting there produces a sample far more likely to score high than Indian adults in general, and it also leaves out people who are not online at all. The 41%…

A fictional study written for practice (Spot the flaw #4).

Fictional study Health · Delhi, India

The study

A research group in Delhi set out to estimate how many Indian adults are addicted to their smartphones. They posted a link to an online questionnaire on a popular digital-detox forum and on the forum's social media pages, and left it open for two months. In total 3,450 adults from 26 states completed a published smartphone addiction scale that has good reliability. Using the scale's recommended cut-off, 41% of respondents were classed as addicted. The team issued a press release headed "41% of Indian adults are addicted to their phones" and called for a national awareness campaign.

What is the flaw in this study?

  • A Underpowered sample
  • B Ceiling effect
  • C Regression to the mean
  • D Selection bias in the sample The flaw
Selection bias in the sample. People who visit a digital-detox forum are, almost by definition, worried about their phone use. Recruiting there produces a sample far more likely to score high than Indian adults in general, and it also leaves out people who are not online at all. The 41% describes these respondents, not the country. A prevalence estimate needs a sample that represents the population, ideally a random household sample, or at least weighting to known population figures. The sample is large, so it is not underpowered; a big biased sample simply gives a precise wrong answer.

The concept: Representative sampling for prevalence estimates

This study is fictional, written for practice. No real study, institution or researcher is described.