Selection Bias: Occurs when some members of the target population are systematically excluded from the sample, leading to a sample that is not representative of the population.
- Identify the target population (all teenagers)
- Identify the actual sample (students at one high school)
- Determine if certain groups are systematically excluded
- Assess how exclusion affects results
All teenagers aged 13-19, regardless of education status or location
Surveying only students at one high school excludes home-schooled, dropouts, working teens, etc.
Home-schooled teens, dropouts, working teens, teens not attending that specific school
Results may overrepresent middle-class teens with access to smartphones provided by parents
This exhibits selection bias because the sample excludes teens not attending high school, potentially skewing results toward higher smartphone usage.
• Representativeness: Sample should mirror population diversity
• Systematic Exclusion: Any consistent exclusion creates bias
• Population Match: Sampling frame should cover entire target population
Response Bias: Occurs when respondents provide answers that are not truthful, often due to social desirability pressure or question wording.
"Do you agree that cheating on exams is wrong?" - loaded with moral judgment
Students may feel pressured to give socially acceptable answers
Public stance may differ from private behavior
Social desirability bias causing inflated "correct" responses
This shows response bias due to social desirability pressure - students give expected moral answers rather than truthful ones.
• Neutral Wording: Questions should be neutral and not lead respondents
• Anonymous Surveys: Reduce social pressure to conform
• Indirect Questions: Consider indirect approaches for sensitive topics
Non-Response Bias: Occurs when individuals who do not respond to a survey differ systematically from those who do respond.
Only 25% responded - 75% did not participate
Busy workers, dissatisfied employees, those with privacy concerns
Those who respond might be more satisfied and have time to participate
Results may overestimate satisfaction due to non-response bias
Non-response bias exists because dissatisfied employees may be less likely to participate, inflating satisfaction results.
• High Response Rate: Aim for >70% to minimize non-response bias
• Follow-up Efforts: Make multiple attempts to increase participation
• Response Analysis: Compare early vs. late responders for differences