Within Subjects Participants Designs All participants serve in

Within Subjects (Participants) Designs All participants serve in each condition of the study. Advantage: Individual Differences are not a Confound. ch 8(1) 1

Types of WS Designs 1) Pre-Post Design with no treatment control group. Used when the purpose is show a change between conditions (i. e. treatment or an outcome) and the order of the conditions is logically set (i. e. , pre test must proceed post test). ch 8(1) 2

2) WS design with counterbalancing for order. Used when comparing conditions and the conditions can be measured in any order. Weakness: Sequencing Effects: Participation in one condition may effects participation in a subsequent condition(s). ch 8(1) 3

A) Order Effects Changes in performance due to practice, familiarity, boredom that are due to repeated testing and not to differences in the IV. ch 8(1) 4

B) Carry-over Effects Long-lasting effects participation in one condition that carry-over to effect performance in subsequent conditions. e. g. , Learning, Expectation, attitude, physical changes due to experimental condition. E. g. , Reward Contrast effects in Rats ch 8(1) 5

Example of Carry Over effects: Reward Contrast effects in Rats ch 8(1) 6

Controlling for Sequence Effects (WS Design) 1) Counterbalance - controls for order effects and for unexpected carry-over effects. Counterbalancing: Within your study, each condition occurs equally as often in each ordinal position, and each condition occurs equally as often before and after each other condition. ch 8(1) 7

Randomized Counterbalancing Each Subject gets a different order. ABC ACB BCA BAC CAB CBA Subjects are randomly assigned to order conditions. ch 8(1) 8

Complete Randomized Counterbalancing All possible orders of conditions are used in and participants are randomly assigned so that there are equal numbers in each order condition. ch 8(1) 9

Equal numbers of subjects are assigned to each order condition, thus order effects are controlled by equalling the effects of order for each condition. Condition A Condition B Condition C 1 st 2 nd 3 rd 1 st 3 rd 2 nd 3 rd 1 st 2 nd 1 st 3 rd 1 st 2 nd ch 8(1) 10

Incomplete Counterbalancing With 4 conditions all orders would total 24. ABCD BACD CABD DABC ABDC BADC CADB DACB ACBD BCAD CBAD DBAC ACDB BCDA CBDA DBCA ADAB BDAC CDAB DCAB ADBA BDCA CDBA DCBA ch 8(1) 11

Rules for Incomplete Counterbalancing 1) Each condition must appear an equal number of times in each ordinal position. 2) Each condition must precede and be followed by every other condition equal numbers of times. A B D C B C A D C D B C D A C B ch 8(1) 12

Intrasubject counterbalancing Each subject gets each possible order. ABCCBA ch 8(1) 13

Between Subjects Designs – Each participant serves in only one condition in the study. Confound: Individual Differences ch 8(1) 14

Control of Individual Differences e. g. , IQ, personality, sex, race, age etc. Random Assignment - Each member of the sample has an equal chance of being assigned to any of the conditions. ch 8(1) 15

Power Ability to find a difference between conditions when a difference actually exists. When designing a study we want high power!! ch 8(1) 16

Power and Designs 1) WS designs have higher power than BS designs and should ALWAYS be considered first. ASK Can I counterbalance for order? If yes, then counterbalance. If no, use pre-post test with no-treatment control group. ch 8(1) 17

BS designs are used only when participants cannot be in more than one condition of the study. • Being in each condition permanently changes a person so that they cannot be in other conditions of the study. ch 8(1) 18

Pre Post Change Cognitive Therapy Drug treatment Combined Treatment No treatment Control Comparison between treatment types here is a BS comparison ch 8(1) 19

BS designs are commonly used for Natural Manipulation studies (e. g. , sex differences, personality differences). These are not experiments and they have low Internal Validity. ch 8(1) 20