Brainscape's Knowledge GenomeTM

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


See full index

psych 200 > Q1 methods & designs > Flashcards

Q1 methods & designs Flashcards

1
Q

design

A

what type of study is the most appropriate?

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

method

A

what tools will be best to measure the phenomena you are exploring?

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

descriptive studies

A

record what is, identify what people are doing on average

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

descriptive studies do NOT

A

measure relationships, make predictions, test hypotheses, or determine causation

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

most studies have a _____ component but follow up with _______ research

A

most studies have a descriptive component but follow up with correlational or experimental research

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

3 types of descriptive research

A

survey, demographics, epidemiological

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

survey

A

provide individuals with questions and they respond

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

demographics

A

report age, race, income level, education level, etc.

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

epidemiological studies

A

report rates or incidents of mental disorders, etc. and how these change across age groups, time, regions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

correlational research

A

describes the relationship between two or more variables
if x increases, what happens to y?
no inference of causation due to third variable problem or spurious correlations

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

can we infer causation from correlational studies?

A

NO

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

what are good practices in correlational research?

A

identify variables of interest and variables which may confound your results
consider measuring something that shouldn’t correlate with your variable
AND something that should be similar to differentiate your proposed explanatory variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

experimental designs

A

allow you to draw inferences about causation
manipulate one independent variable
assign participants to a condition with 1 level of the independent variable
control for confounding variables
pilot testing, manipulation check, blinding, controls, etc.

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

conditions

A

the total number of groups within an experiment

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

levels

A

different values of the independent variables (qualitative or quantitative)
if you have two variables you need one level to be a negative control

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

control condition

A

ensure that the observed effect is truly due to the manipulation; allows for a baseline

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

types of independent variables (3)

A

environmental, instructional, invasive manipulations

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

environmental manipulations

A

confederates, actors in your experiment

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

instructional manipulations

A

change what you tell them to do or how, wording

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

invasive manipulations

A

give people things that may stimulate the nervous system, treatments, etc.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

between groups design

A

randomly assign participants to a group to ensure equivalence (equal chance to be in any condition) and measure differences between the groups

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

within groups design

A

randomly determine what order participants receive all of your conditions; compare results within individuals subjects in the different conditions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

order effects

A

the order of conditions might impact the results of the study; to avoid this, we randomize in which order the participants complete the tasks

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

advantages of within groups design

A

more power (ability to detect an effect) since individuals do not differ, need smaller samples than between subjects designs, less noise

25
Q

importance of random assignment

A

allows us to assume any differences in behavior are due to your independent variable and not confounding variables

26
Q

confounding variables

A

uncontrolled extraneous variables that varies systematically other than the independent variable; the more you control for in your study, the more participants you need

27
Q

pilot testing

A

ensures that the manipulation was successful and your study words; helps to rule out potential explanations in null results

28
Q

manipulation checks

A

ensure that the manipulation is effective and that your participants are perceiving the stimuli/manipulation like you want them to

29
Q

subject variables

A

variables that may vary across participants but cannot be manipulated by the experimenter (gender, IQ, socioeconomic status, personality traits, childhood characteristics)

30
Q

internal validity

A

certainty that the independent variable caused the change in the dependent variable
lowered when you have systematic variation not explained by the independent variable (confounds)

31
Q

differential attrition

A

if a certain subsection of participants from your study seem to drop out at a higher rate

32
Q

quasi-experimental design

A

examining differences between two different groups without random assignment to those groups
cannot draw causal inferences in these cases since you do not manipulate a variable but instead use already existing groups

33
Q

experimenter/expectancy effects

A

observers interpret behavior in a biased way (usually based on the hypothesis for the experiment)

34
Q

demand characteristics

A

participants pick up what you’re trying to explore in your study and may change their behavior or responses based on what they think you want, as opposed to their normal behavior

35
Q

placebo effect

A

people experience change in their behavior or improvement because they think something will happen
helps you get a sense of whether the behavior or treatment would improve if there was an ineffective treatment (they think they’re getting one, but it has no effect)

36
Q

double blind procedure

A

golden standard for medical treatments
neither the participants nor researchers know which participants are receiving which treatment
works to prevent expectancy and demand characteristics effects

37
Q

external validity

A

can i generalize my results beyond the lab?

38
Q

experimenter’s dilemma

A

you cannot always have great internal and external validity
if you control for all confounds (internal validity), your results may not be as applicable outside of the lab
thus, there is an inverse relationship between internal and external validity

39
Q

do experiments prefer internal or external validity?

A

internal validity
you want to be able to attribute variation in your dependent variable to your manipulation of the independent variable

40
Q

experi-corr designs

A

experimental and correlational research combined
independent variable is manipulated and subject variables are measured
do any variables moderate/qualify the effect of the independent variable?
how do personal attributes relate to outcomes under varying conditions? could certain variables predict (correlation) the causal relationship between two variables? (experimental)

41
Q

examples of methods (4)

A
  1. interview
  2. surveys
  3. psychphysiology/involuntary
  4. observation
42
Q

issues in surveys

A

response bias
unrepresentative samples

43
Q

response bias

A

willingness or unwillingness to answer a survey based on what is going on in someone’s life

44
Q

questionnaire

A

validated measure (measuring what it is intended to), run statistics to make sure it is reliable (same results if taken 3 days apart), prior testing

45
Q

population

A

we want to make generalizations about a given population, but we cannot test all of the people in a given population
this is the entire group about whom the research question is asked

46
Q

sample

A

the smaller subset of the population that you include in your study; hopefully this sample is representative of the population

47
Q

sample error

A

sometimes your sample may not exactly capture the make up or reflect the behavior of the population

48
Q

inferential statistics

A

generalize from a sample to a population; infer the population’s behavior/data based on the sample

49
Q

null hypothesis

A

assumes no difference between groups or no association between variables

50
Q

alternative hypothesis

A

there is a difference between groups or an association between variables

51
Q

4 outcomes for hypothesis testing

A
  1. reject the null hypothesis when there is an effect (correct)
  2. fail to reject the null hypothesis when there is NOT an effect (correct)
  3. reject the null hypothesis when there is NOT an effect (type i error)
  4. fail to reject the null hypothesis when there is an effect (type ii error)
52
Q

type i error

A

rejecting the null hypothesis when there is NOT an effect (alpha)

53
Q

type ii error

A

failing to reject the null hypothesis when there IS an effect (beta)

54
Q

power of a test

A

1 - beta (probability of making a type ii error)
the probability you reject the null hypothesis correctly

55
Q

ideal power of a test

A

0.8, or an 80% chance of detecting an effect

56
Q

how can you increase the power of a test?

A

decrease standard deviation
increase sample size
increase alpha level
increase the difference between the null and alternative hypotheses

57
Q

ideal beta value

A

0.2, or a 20% chance of not detecting a given effect (type ii error)

58
Q

what kind of relationship do alpha and beta have?

A

inverse, as one increases, the other decreases

59
Q

why do we use alpha to determine if a result is significant?

A

a type i error is more impactful

psych 200 (15 decks)

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