Conducting Longitudinal Studies and Randomised Controlled Trials (RCTs) in Psychology and Education

This course will give an overview of issues in the design, implementation, analysis and reporting of longitudinal studies and RCTs in psychology and education. The course will not include any “hands on” data analysis, but it will include discussion of key statistical/methodological issues.

Professor Charles Hulme

Learning outcomes

This course is designed to be useful to students and early career researchers who are considering conducting longitudinal or intervention studies in the fields of psychology and education. A key conceptual issue is causation. How do we identify possible causes?– By doing longitudinal studies! Causal theories can lead to recommendations for treatments of disorders of development. Ideally, treatments should be based on a sound theory of the nature of a disorder.The course will make clear why RCTs are so important in evaluating proposed treatments. RCTs are also critical for evaluating the causal theories that underlie them.

Admission

Ph.d-students at The Faculty of Educational Sciences will be given priority, but the course is also open for others.

Registration: Please use the registration form

Deadline for registration: 21 August 2017 

Credits: No credits

Teaching

This is a two day course (part I and part II) organized with 2-3 hours sessions, each consisting of a lecture followed by a class discussion. You can sign up for both days or one day only.

Part I, 29 August: Conducting Longitudinal Studies

Part II, 30 August: Randomised Controlled Trials (RCTs)

Organiser: Department of Special Needs Education

Responsible and lecturer: Prof. Charles Hulme, University of Oxford, Prof. II Department of Special Needs Education, University of Oslo

Dates and hours: 29-30 August, 10:00-15:00 both days.

Room: 231, Helga Engs house

Language: English

Part I: Longitudinal studies

Two sessions to cover:

  • Planning and conducting a longitudinal study. 
  • Issues of applying theory to planning an effective longitudinal study. 
  • Selecting measures – issues of measurement (reliability and validity)
  • Writing a paper reporting a longitudinal study. 
  • Data analytic techniques – regression models, structural equation models (especially autorergressive path models and growth curve models). 

Examples will come from Professor Hulmes own longitudinal studies of reading and language development. 

Key references

Muter, V., Hulme, C., Snowling, M.J., & Stevenson, J. (2004).  Phonemes, rimes and language skills as foundations of early reading development: Evidence from a longitudinal study.  Developmental Psychology, 40, 663-681.

Lervåg A., Bråten, I., and Hulme, C.  (2009) The cognitive and linguistic foundations of early reading development:  A Norwegian latent variable longitudinal study.  Developmental Psychology, 45, 764-781.

Caravolas, M., Lervåg, A., Mousikou, P., Efrim, C., Litavský, M., Onochie-Quintanilla, E., Salas, N., Schöffelová, M., Defior,S.,  Mikulajová, M., Seidlová-Málková, G., Hulme, C.  (2012)  Common patterns of prediction of literacy development in different alphabetic orthographies. Psychological Science, 23(6), 678-86.

Goebel, S., Watson, S.  Lervag, A, & Hulme, C.  (2014)  Children's arithmetic development: it is number knowledge, not the approximate number sense,  that counts.  Psychological Science, 25, 789-798.

Hulme, C., Nash, H.  Gooch, D., Lervag, A., Snowling, M.J.  (2015)  The foundations of literacy development in children at risk of dyslexia.  Psychological Science, 26,  1877-1886.  DOI: 10.1177/0956797615603702

Malone, S., Burgoyne, K., Tolan, A. & Hulme, C.  (in press ) The cognitive and sensori-motor foundations of arithmetic: Finger gnosis predicts variations in counting and number knowledge in young children.  Developmental Science

Part II: Randomised Controlled Trials (RCTs)

Two sessions to cover:

  • Planning and conducting an RCT
  • The logic of random assignment. 
  • Allocation of participants to groups (randomisation and minimisation). 
  • Identifying key outcome variables and how to measure them 
  • Statistical power and why adequate group sizes are critical. 
  • The use of Analysis of Covariance models and why they are the preferred analytic technique for RCTs. 
  • Conducting and interpreting mediation and moderation analyses. 
  • Reporting a trial in accordance with the CONSORT guidelines.

Examples will come from Professor Hulmes own longitudinal studies of reading and language development. 

Key references

Bowyer-Crane, C., Snowling, M.J., Duff, F.J.  Fieldsend, E.  Carroll, J., Miles, J.N.V., Goetz, K., Hulme, C.  (2008). Improving early language and literacy skills: Differential effects of an oral language versus a phonology with reading intervention. Journal of Child Psychology & Psychiatry, 49, 422-432

Hulme, C., Bowyer-Crane, C., Carroll, J., Duff, F., & Snowling, M.J. (2012). The causal role of phoneme awareness and letter-sound knowledge in learning to read: combining intervention studies with mediation analyses.  Psychological Science, 23(6), 572-7

Clarke, P., Snowling, M., Truelove, E.  & Hulme, C. (2010). Ameliorating children’s reading comprehension difficulties: A randomised controlled trial. Psychological Science, 21, 1106-1116

Fricke, S., Bowyer-Crane, C., Haley, A. Hulme, C., & Snowling, M.  (2013)  Efficacy of language intervention in the early years. Journal of Child Psychology and Psychiatry, 54, 280–290. 

Senn, S. (2006) Change from baseline and analysis of covariance revisited. Statistics in Medicine, 2006; 25:4334–4344

Bland, M & Altmann, D. (2005). Treatment allocation by minimization. British Medical Journal, 330, 843.

Vickers, A. & Altmann, D. (2005). Analysing controlled trials with baseline and follow up measurements. British Medical Journal, 323, 1123 1124.

 

 

Publisert 2. juni 2017 15:21 - Sist endret 14. aug. 2017 15:15