Randomised controlled trial of incentives to improve attendance

  • Slides: 11
Download presentation
Randomised controlled trial of incentives to improve attendance at adult literacy classes Greg Brooks*,

Randomised controlled trial of incentives to improve attendance at adult literacy classes Greg Brooks*, Maxine Burton*, Pam Cole*, Jeremy Miles**, Carole Torgerson*** and David Torgerson** g. brooks@shef. ac. uk *School of Education, University of Sheffield **York Trials Unit, Dept of Health Sciences, University of York ***Department of Educational Studies, University of York

Background • Poor adult literacy widespread problem • Regular attendance known to correlate with

Background • Poor adult literacy widespread problem • Regular attendance known to correlate with adult learners making better progress in reading. • No evidence from randomised controlled trials on incentives to attend.

Methods • Pragmatic cluster randomised controlled trial • Built on top of existing University

Methods • Pragmatic cluster randomised controlled trial • Built on top of existing University of Sheffield Effective practice in reading study

Funding for main study Funding for trial ESF Df. ES Sf. LSU Io. E

Funding for main study Funding for trial ESF Df. ES Sf. LSU Io. E NRDC University of Sheffield (separate fund) Learners

 • Ethical approval and informed consent obtained • 29 adult literacy classes allocated

• Ethical approval and informed consent obtained • 29 adult literacy classes allocated by York Trials Unit to two groups using minimisation • Classes mainly in East Midlands and North of England, with 3 outliers in South East • Main outcome: number of sessions attended • Secondary outcome: reading attainment (tests devised by NFER)

 • Both groups received £ 20 to attend preand post-tests sessions in January

• Both groups received £ 20 to attend preand post-tests sessions in January and June 2005 • Intervention group also received £ 5 for each session attended between pre- and post-tests • Payments to learners after end of study

Results • One class did not meet inclusion criteria – excluded • 14 classes

Results • One class did not meet inclusion criteria – excluded • 14 classes in each group • Tests marked by researcher at NFER • Data analysed by statistician in York blind to status of groups

Effects of incentives on sessions attended and post-test scores Variable Intervention (n = 82)

Effects of incentives on sessions attended and post-test scores Variable Intervention (n = 82) Control (n = 70) Mean (SD) number of sessions attended (p = 0. 019) 5. 28 (2. 79) 6. 69 (2. 71) Mean (SD) post-test literacy scores (not significantly different from pre-test for either group) 19. 01 (8. 68) 21. 14 (8. 84)

Limitations • Small study • Small incentive • Incentives in form of vouchers –

Limitations • Small study • Small incentive • Incentives in form of vouchers – cash better? • Did not test policy of financial sanctions

Discussion Perverse result known variously as: • ‘Over-Justification Hypothesis’ • ‘Corruption Effect’ • ‘The

Discussion Perverse result known variously as: • ‘Over-Justification Hypothesis’ • ‘Corruption Effect’ • ‘The Hidden Cost of Reward’ • ‘Cognitive Evaluation Theory’ • ‘Crowding-Out Effect’

External interventions crowd out intrinsic motivation if they are perceived as controlling. In that

External interventions crowd out intrinsic motivation if they are perceived as controlling. In that case, both selfdetermination and self-esteem suffer, and the individuals react by reducing their intrinsic motivation in the activity controlled. (Frey and Jegen, 2001, p. 594) • - with rider ‘for interesting tasks’ added on p. 598 • Plausible explanation?