Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

EDUC222 > Week 3: Validity part 1 > Flashcards

Week 3: Validity part 1 Flashcards

1
Q

validity

A

approximation to truth/falsity of propositions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

4 types

A
  • statistical conclusion
  • internal
  • construct
  • external
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

statistical conclusion

A

covariation of variables: changes in A detectable changes in B (do we see an effect on B)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

threats to statistical conclusion (list) (7)

A
  • low statistical power (type 2 error)
  • violated assumptions on tests
  • fishing
  • measurement reliability
  • random irrelevancies in experimental setting
  • random heterogeneity of respondents
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

type 2 error

A
  • sample size/n is small - alpha is low
  • two-tail hypothesis
  • manipulation wasn’t strong enough
  • w/i group variance is large
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

violated assumptions on test

A
  • test if assumptions are true before coming to conclusions about covariance
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

fishing/type 1 error

A
  • multiple comparisons increase chance

- cure: stats

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

measurement reliability

A
  • if unreliable -> cannot be counted

- cure: have multiple items in test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

treatment implementation reliability

A
  • lack of standardization
  • can inflate variance
  • cure: consistency/standardized
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

random irrelevancies in experimental setting

A
  • some features of setting affect DV scores -> inflate error variance
  • monitor experiment closely
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

random heterogeneity of respondents

A
  • people differ in qualities/sensitivity to IV/DV/measurements
  • homogenous sample
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

internal validity

A
  • causal relationship (do we know A caused B)
  • time sequence
  • not easily established
  • threats depending on design
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

pre-post test design

A
  • pre-test -> manipulation -> post tests

- comparing scores before/after manipulation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

threats (pre-post test design w/no random assignment)

A
  • history
  • maturation
  • testing
  • instrumentation
  • statistical regression
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

history

A
  • effect bc event took place between pretest and post test

- treatment: controlled environment, DV unaffected by other things

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

maturation

A
  • participants grow older/wiser/stronger/more experienced between pretest and post test
17
Q

testing

A
  • pre-test may cause change in DV (not experimental manipulation
18
Q

instrumentation

A
  • measurment instrument becomes different from pre-test to post test
  • eg interviewers becoming skilled
19
Q

instrumentation: ceiling vs floor effects

A

c: test too easy, scores all high
f: test too difficult

20
Q

statistical regression/regression to the mean

A
  • participant selection based on extreme scores
  • probability of scoring high/low again is low
  • treatment: matching (control group w/same number of participants w. extreme scores
21
Q

internal validity threats (between subjects no randomization design)

A
  • selection
  • mortality
  • interaction w/selection
22
Q

selection

A 
- w/o random assignment: effect on DV can be bc of systematical differences between two groups 
(eg advanced class vs regular) 
- treatment: random assignment
23
Q

mortality

A
  • ppl drop out before study is over
  • differential mortality (one group drops out -> random assignment to fix)
  • random mortality (no treatment)
24
Q

statistical interactions w selection

A
  • selection-maturation
  • selection-history (local history)
  • selection-instrumentation
25
Q

selection-maturation

A

participants in one group mature at a different rate than another group

26
Q

selection-history (local history)

A

something happens in one group but not the other

27
Q

selection-instrumentation

A

one group’s characterstic interacts w/ experimental intervention (eg underachievers vs A+ students)

28
Q

random assignment

A

each participant has equal chance to be in any of the GROUPS in a study (NOT RANDOM SAMPLING FROM A POPULATION)

29
Q

internal validity threats (between subjects w random assignment)

A
  • local history
  • diffusion of treatment to control group
  • resentful demoralization of control group
  • compensatory equalization of treatment
  • compensatory rivalry by respondent receiving less desirable treatments
30
Q

local history

A
  • something happens only in one experimental condition

- eg: 2-session study about memory (one recalls out loud)

31
Q

diffusion of treatment to control group

A

info about treatment goes from one group to the other

- keep them from interacting

32
Q

control group’s resentful demoralization

A
  • control group receiving less favorable treatment

- keep them from interacting

33
Q

compensatory equalization of treatment

A

experimental manipulation yields gains (eg goods) -> reluctance to participate in control group

34
Q

compensatory rivalry

A
  • group assignment becomes public -> social competition
  • control group motivated to reduce/revers expected difference between groups
  • keep them from talking
  • don’t reveal info on assignment
35
Q

construct validity

A
  • adequacy of operational definition

- causal relationsihp in theoretical structures

36
Q

external validity

A

generalizing relationship (can this happen in other situations/the natural real world)

37
Q

experimentation vs reality (e/r)

A
  • t/t: correct
  • f/t: type 2 false negative
  • t/f: type 1 false positive
  • f/f: correct

EDUC222 (6 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions