PSY 250 Chapter 7 Experimental Research Strategy Cause

PSY 250 Chapter 7: Experimental Research Strategy

Cause and Effect Relationships n Can be clear relationship but not a causal one – e. g. between getting dressed up and having a headache the next morning n 4 basic elements to establish cause and effect: 1. 2. 3. 4. Manipulation Measurement Comparison Control

Variables n Independent n Dependent n Extraneous n Levels of the IV n The different values of the IV selected to create and define the treatment conditions n Must be ≥ 2 n Treatment Condition n Situation or environment characterized by one specific value of the IV

The Third-Variable Problem n Can be clear relationship but not a direct causal one between 2 variables n E. g. children’s involvement in extra-curricular abilities and confidence n 3 rd variable may be controlling both of the other 2 (family involvement, opportunities for more friendships) n Research becomes the art of teasing apart and separating a set of naturally interconnected variables

Control n Eliminate all confounding variables n E. g. Meaningfulness and imagery (Paivio, 1965) n Create two lists (high vs. low imagery) with equal average meaningfulness n Classification of confounding vs. independent variable depends on research hypothesis

The Directionality Problem n Which is the cause and which is the effect n E. g. assertiveness and success n Sleep and depression

Manipulation and Control n Unique to experimental research strategy n Manipulation n Create treatment conditions corresponding to values of IV n If change in A does not cause change in B then A is not a causal agent (unless change is too small to affect B)

Control n Prevent extraneous variables from becoming confounding variables n Only confounding if it influences DV AND n It varies systematically with IV n 3 types of extraneous variables: Environmental n Participant n Time - Related n

Methods of Controlling Extraneous Variables n 1. Holding a Variable Constant n Hold absolutely constant or n Limit to restricted range But can limit generalizability n What is this a threat to? n

Methods of Controlling Extraneous Variables n 2. Matching Values across Treatment Conditions E. g. each condition has = males and females n Or ensure average value is same – e. g. average is 4. 5 in each group n Also match on environmental and time variables n

3. Matching by Yoked Control n Controls for possible influence of temporal relationship between event and response n Ulcers due to physical or psychological stress of shock in monkeys (Brady, 1958)

4. Building Extraneous Variable into the Design n. Sample of participants n with IQs of 90 - 119 n Subsamples with nspecified IQ values identified n n. Participants nwith 90 -99 IQ n. Participants nwith 100 -109 IQ n IQ values nbuilt into design n 90 - 99 n 100 -109 n 110 -119 n. Learning n. A n strategy n. B n. Participants nwith 110 -119 IQ

5. Matching by Equating Participants n Also increases sensitivity of exp. n Similar to building EV into design n # of participants is some multiple of the # of levels of the IV n Two methods

5 a. Precision Control n. Sample of research n participants n. Each participant measured in nterms of age, gender, IQ, etc n. Each pair of matched participants n n randomly assigned to the treatment conditions Treatment A Treatment B n Each participant matched nwith 2 nd participant of same age, gender, IQ, etc. n n. Provides many matched n pairs of participants

5 b. Frequency Distribution Control n 1 st sample of n participants n. Mean = 107 n S. D. = 5. 4 n. Similar frequency n distribution of n IQ scores n 2 nd sample of participants n n. Mean = 106. 5 n S. D. = 5. 1

Methods of Controlling Extraneous Variables n 6. Randomization n Disrupts any systematic relation between EVs and IV – prevents EVs from becoming CVs n Unpredictable, unbiased procedure to distribute different values of each EV across treatment conditions n All possible outcomes equally likely n But chance CAN produce biased outcomes – e. g. all heads with 10 coin tosses

Control Through Participant Assignment n Random assignment n ensures influential extraneous variables are balanced among experimental conditions n “whenever possible, randomize” n representative n Free random assignment n random numbers

Counterbalancing n To control for sequencing effects n Order effects n IV – rate of presentation of nonsense syllables n DV – verbal learning n Learn slow, moderate then fast list – speed confounded with order n Carry-over effects n Performance in condition partially dependent on preceding conditions n IV – monetary reward n Dime may be more rewarding when preceded by 5 vs. 15 cents

Control Groups n No treatment condition n Provides baseline measure of normal behavior n Experimental group – treatment condition n Placebo control groups

General Control Procedures n Preparation of setting n structuring research setting n Response measurement n careful selection and preparation of measures n Replication n systematic and conceptual replication

Control Over Subject and Experimenter Effects n Experimenter effects n Participant (subject) effects n demand characteristics

Control Over Subject and Experimenter Effects (cont’d) n Single-blind procedures n experimenter is unaware n Partial-blind n Experimenter unaware for portions of exp. n Double-blind procedures n experimenter and participant are unaware n Deception n Participant is unaware

Control Over Experimenter Effects (cont’d) n Automation n reduces experimenter-participant contact n Using objective measures n require minimal judgements

Control Over Subject and Experimenter Effects (cont’d) n Multiple observers n ratings of behaviors n Control of experimenter attribute errors n physical and psychological characteristics n Hold constant across treatments

Gaining Insight Into Participants’ Perceptions of Experiment n Experimental manipulation check n Did manipulation work? n What did participant think experiment was about? n How does participant think others will respond? n Participant Manipulations n Subtle Manipulations n Simulations n Placebo Controls

Simulation Studies n Try to duplicate natural environment in lab – bring real world into lab n Mundane Realism n Superficial, physical characteristics n Experimental Realism n Psychological aspects of situation n Do participants become immersed in situation n E. g. Haney, Banks & Zimbardo (1973) prisoner study n Prisoner’s dilemma

Field Studies n Bring lab into real world n E. g. Bystander Apathy in emergency situations