Developmental Research Methods Topic 7: Control 06/07/2022 CEDP321 Ryan Sain, Ph.D. 1
Dec 21, 2014
04/10/2023 CEDP321 Ryan Sain, Ph.D. 1
Developmental Research MethodsTopic 7: Control
04/10/2023 CEDP321 Ryan Sain, Ph.D. 2
Control through selectionFirst step in making our
experiment internally valid Selection
◦Who to select?◦How?◦Must be representative
04/10/2023 CEDP321 Ryan Sain, Ph.D. 3
Simple Random Sampling Population Randomly select Fair? Representative
04/10/2023 CEDP321 Ryan Sain, Ph.D. 4
Random assignmentBalances differences that we
didn’t control for through selection.
This minimizes confounds – or at least makes the equally likely to occur in both groups
But it doesn’t get rid of them◦It’s entirely possible that all females
end up in one group◦Just makes it unlikely
04/10/2023 CEDP321 Ryan Sain, Ph.D. 5
Experimental design controlDesign allows us to detect causeWe are all about ruling out
confounds.Not all designs created equal
04/10/2023 CEDP321 Ryan Sain, Ph.D. 6
True ExperimentsTrue experiments
◦Random selection and random assignment
◦Controls for most threats to internal validity History, maturation,
selection, testing, etc.
04/10/2023 CEDP321 Ryan Sain, Ph.D. 7
Selection process
Group Measurement IV level Measurement
- Group A - Treatment DV
No ability to detect cause
(r) Experimental - Treatment DV
(r) Control - No Treatment DV
Starting to detect cause, but we’re not sure about natural change (only one measure) and random effectsR Experimental Pretest Treatment Posttest
R Control Pretest No Treatment posttest
Now we are able to start detecting changes within each group over time (2dv) to allow for measurement of maturation effectsR Exp 1 Pretest Treatment Posttest
R Control 1 Pretest posttest
R Exp 2 Treatment posttest
R Control 2 Posttest
The solomon 4 group – rules out maturation, testing, and selection effects
04/10/2023 CEDP321 Ryan Sain, Ph.D. 8
Creditswww.sxc.hu
◦Gears◦Beakers – 123dan321◦Puzzle – lusi
Minions – Illumination Entertainment
People – Ryan Sain