Non-Experimental Methods

Question 1

how can we turn survey research into an experiment

Answer 1

asking people different versions of a question instead of preparing one type of questionnaire

Question 2

non-response bias

Answer 2

a concern when survey responder differ in important ways from non-responders

Question 3

leading question

Answer 3

wording that leads to a particular response
example:
how much do you support the gun control policy?

Question 4

double barreled questions

Answer 4

asking two or more questions in one
- fatiguing to answer
example: how do you feel about the biden administration? do you think he is doing enough to solve the homelessness crisis?

Question 5

negative wording

Answer 5

negative phrasing (not, never) is confusing

Question 6

item order effect

Answer 6

when the order in which questions are asked influences participants responses
example:
how is your love life? how happy do you feel right now?
how happy do you feel right now? how is your love life?

Question 7

yea saying

Answer 7

bias towards saying yes to any item

include reverse worded questions

Question 8

fence sitting

Answer 8

answering in the middle of a scale

take away neutral response option

Question 9

socially desierable responding

Answer 9

sometimes participants answer based on the highest version of themselves or they wanna appear good in front of others

assure anonymity; include questions that detect socially desirable responding

Question 10

limitations of introspection

Answer 10

sometimes we are not good at bring truthful
know the limits of self reported data

Question 11

bivariate correlation

Answer 11

correlation coefficient: measure the strength of the relationship between two variables
large numbers: stronger relationship
positive numbers: vary in same direction
negative number: vari in opposite direction

.10 small (weak)
.30 medium
.50 large (strong)

closest to zero - weakest
closest to one - strongest

Question 12

how is statistical control useful?

Answer 12

statistical control helps improve internal validity but cannot establish causation
helps deal with confounds

Question 13

multiple regression

Answer 13

statistical technique to rule out third variables
multiple regression asks the

question: when you take the relationship between the third variable and the DV into account, is there still a portion of the variability in DV that is attributable to the IV?

example: when you take wealth and longevity in account, is there still a portion of the variability in longevity that is attributable to drinking fine wine?

Question 14

moderation analysis

Answer 14

• asses whether a third variable CHANGES the relationship between X and Y
• identified the INTERACTIONS among predictors

Question 15

special types of multiple regression

Answer 15

moderation
mediation

Question 16

mediation analysis

Answer 16

• asseses whether a third variable EXPLAINS the relationship between X and Y
• identified possible causal MECHANISMS

Question 17

moderator

Answer 17

variable that changes the relationship between two other variables
- correlational relationship
- statistical interaction

are there certain groups or situations for which two variables are more strongly related?

Question 18

third variable problem

Answer 18

two variables are correlated because they are both linked to a third variable

Question 18

mediator

Answer 18

why are two variables related?
mediator

IV —- X—-> DV