Selectivity of attrition in longitudinal studies of cognitive functioning

Area of Science:

• Psychology
• Gerontology
• Longitudinal Studies

Background:

• Longitudinal studies are crucial for understanding developmental changes across the lifespan.
• Selective attrition, where participants drop out, can potentially bias study results and limit generalizability.
• Understanding the characteristics of participants who continue versus those who drop out is essential for interpreting longitudinal data.

Purpose of the Study:

• To identify characteristics that differentiate participants who continue in a longitudinal study from those who do not.
• To assess whether longitudinal changes observed in continuing participants accurately represent changes that would have occurred if all initial participants had remained.

Main Methods:

• Analysis of data from a large sample (N=3,780) of returning and nonreturning participants aged 18-97.
• Participants completed cognitive tests, personality, and mood questionnaires.
• Multiple regression analyses were used to examine differences based on age, returner status, and their interaction.

Main Results:

• Returning participants, at baseline, exhibited higher cognitive abilities and specific personality traits (e.g., agreeableness, openness) compared to nonreturning participants.
• These differences were more pronounced in adults over 50 years of age.
• Crucially, no evidence suggested that the longitudinal changes in nonreturning participants would differ from those who returned.

Conclusions:

• Selective attrition in longitudinal research is more intricate than commonly believed.
• The findings suggest that selective attrition may not inherently compromise the generalizability of longitudinal comparisons.
• Researchers should consider participant characteristics but not assume generalizability is lost due to attrition.