Chapter 4

Question 1

types of studies

Answer 1

• experimental designs
• non-experimental designs
• quasi-experimental designs

Question 2

non-experimental designs

Answer 2

• relationships studied by simply observing/measuring variables
• ex. Observation, Correlation

Question 3

correlation

Answer 3

• indexes the degree of LINEAR relationship between two variables
• AKA: Pearson product moment correlation
• Correlation coefficient: r
• Ranges from –1 to +1 (-1 is perfect negative linear relationship, +1 is perfect positive linear relationship,
  0 is no relationship)

Question 4

Why can’t we assume that correlation equals causation?

Answer 4

• Bidirectionality problem: A may cause B, but B may be causing A
• Third variable (or extraneous variable) problem: C may be affecting both A and B
• correlational studies lack internal validity

Question 5

Internal validity (and what is needed to achieve it)

Answer 5

• the ability to infer that one variable causes changes in another variable
• For internal validity, 3 things are needed:
  1) Covariation between two variables (when one variable is present, so is the other one. When one variable isn’t present, neither is the other one)
  2) temporal precedence (if you’re suggesting that A causes B, A has to come before B)
  3) eliminate alternative explanations (ex. Third variable problems, etc.)

Question 6

experimental designs

Answer 6

• directly manipulating and controlling variables
• Allow for causality due to higher internal validity
• Achieve internal validity through experimental control and random assignment

Question 7

2 kinds of experimental designs

Answer 7

• Between subjects

- Within subjects

Question 8

experimental control

Answer 8

• treating participants in all groups identically so that the only difference between the groups is the independent variable
• minimizes confounds by ensuring that only the IV changes across conditions

Question 9

confound

Answer 9

• A variable that co-varies along with the IV
• Could explain all or part of the result
• Difference between confound and third variable: Confounds are present in experimental studies; third variables are present in correlational studies
• Can combat confounds using solid operational definitions

Question 10

random assignment

Answer 10

• increases internal validity
• Allows researchers to balance out random variables across different conditions – on average
• More effective as the number of participants increase

Question 11

Operational definition

Answer 11

defines the variable in terms of the concrete operations or techniques used to measure or manipulate it in a specific study

Question 12

3 kinds of operational definitions

Answer 12

1. situational variable
2. participant variable
3. response variable

Question 13

situational variable

Answer 13

• describes characteristics of situation or environment
• Ex. The length of words read, the number of bystanders present, etc.
• Can be measured in any study and manipulated in experimental studies

Question 14

response variable

Answer 14

• responses or behaviours of individuals
• Ex. Reaction time, performance on cognitive task
• Can be measured in any type of study

Question 15

participant variable

Answer 15

• characteristic that individuals bring with them to a study
• Ex. Cultural background, intelligence, personality

Question 16

curvilinear relationship

Answer 16

increases in one variable accompanied by both increases and decreases in the other variable

Question 17

mediating variable

Answer 17

• something that links the IV and DV and helps explain the relationship between them
• Ex. Punching a child causes a child to cry. The mediating variable in this case is pain.

Question 18

lab studies

Answer 18

• allow variable to be manipulated in carefully controlled confines of a lab, reducing possibility that extraneous variables could influence the results.
• this may also create an artificial atmosphere that may limit generality of results

Question 19

field experiment

Answer 19

variables are manipulated in a natural setting (such as street corners, malls, online, etc.). Less artificial context, but researcher loses ability to directly control many aspects of the situation.