research methods Flashcards
1
Q
internal validity
A
- how much we can trust the findings of the study are due to the factors explored in the study
- confounding variables
- control groups and randomization
- clinical sample vs. non clinical sample vs. analogue sample
2
Q
confounding variables
A
- factors that were not explored in the study but may have affected the outcome of the study
- ie. age in a computer assisted therapy and depression study
3
Q
control groups and randomization
A
- control groups: where you randomly assign people to a group to serve as a comparison
- randomization: ensuring that the two groups being studied are not systematically different in any way
4
Q
clinical samples
A
- you would potentially diagnose someone as part of the study
- need a qualified person for this
5
Q
non clinical samples
A
everyday people like graduate students
6
Q
analogue sample
A
- an experimental design where the procedures/participants used are similar but not identical to the situation of interest
- ie. if researchers were interested in determining the significance of therapist gender on clients trusting their therapist, it’s hard to get real world data, so you may have graduates read counselling scenarios where the therapist gender would vary in the scripts, then they’d rate how much they think they’d trust them
7
Q
external validity
A
- how much the results of a study are generalizable to the real world
8
Q
statistical significance
A
- an association or difference between variables that was not caused solely by normal variation or chance
9
Q
completer analyses vs intent-to-treat analyses
A
- completer analyses: eliminating all dropouts in a study
- intent-to-treat analyses: assuming that the person did not change since they dropped out
10
Q
effect sizes
A
- interpretation of correlations
- 0.1-0.3 = no correlation
- 0.3-0.5 = moderate correlation
- 0.5+ = strong correlation
11
Q
clinical significance
A
- when a statistically significant finding is substantial enough to be clinically meaningful and thus should direct the course of treatment
12
Q
cross-sectional designs
A
- collects data at a one time point
- ie. brown and finn
- cohort effects
13
Q
brown and finn
A
- wanted to study the factors that seem to correlate with alcohol use
- studies showed the beliefs in alcohol of 12, 15 and 17 year olds
- 12 year olds: 36% of them thought that the point of drinking was to to get drunk
- 15 year olds: 64% held the belief that the point of drinking was to get drunk
- 17 year olds: the percentage dropped to 42%
14
Q
cohort effects
A
when a group of people shares experiences (ie. age and culture) that shape their behavior or attitudes in a way that’s different from other groups
15
Q
longitudinal designs
A
- collects data at multiple time points
- disadvantage: expensive, rare
- advantage: we can determine if a variable actually predicts another
- nagin and tremblay
- cross-generational effects
16
Q
nagin and tremblay
A
- wanted to explore factors that could cause aggression in boys
- sample was 100 boys from when they were 6 to 15 years old (low economic status)
- grouped into 4 levels of aggression: chronic, high but declining, moderate but declining, low aggression
- the first 2 groups were predicted by hyperactive, having a teenage mom, or a mom with less education
17
Q
cross-generational effects
A
- cohort effects across generations
18
Q
correlational research
A
- epidemiological research
- prevalence
- incidence
19
Q
epidemiological research
A
- correlational research that involves studying the prevalence, distribution and consequences of disorders in populations
- massive studies
- gathers thousands of members of the general public for face-to-face interviews
20
Q
prevalence
A
- % of the population who has the disorder
- you can have 12 month prevalences (% of people who have had it in the past year), life-time prevalence (the % of the amount of people who have had in their their lifetime)
21
Q
incidence
A
- number of new cases in a specific time, typically a year
22
Q
experimental research: treatment outcome research
A
- open-label trials
- randomized control trials
- single-blind vs double blind studies
- waitlist control studies
- placebo control studies
- comparison studies
- process studies
- community-based studies
23
Q
open label trials
A
everyone knows the treatment they’re getting
24
Q
randomized control trials
A
- participants sign up and don’t know what they’re signing up for
- assigned to one of the conditions
25
single blind studies
researcher knows what treatment participants are getting, but the participants do not
26
double blind studies
- neither the researcher nor the participant knows what treatment they're in
- the assessors (who is evaluating) are blind to what treatment and who
27
waitlist control studies
- group of participants that sit on a waitlist for the entire duration as the people getting treatment
- so they serve as a control and often provided treatment afterwards
28
placebo control studies
- studies looking at drugs vs placebos
- participants think they're getting the drug but they're getting placebos
29
comparison studies
- having people take different therapies
- group a gets this, group b gets this, group c gets this
30
process studies
look at "why" does it work, breaking down the components to see what parts of treatment help
31
community-based studies
seeing if the therapy applies to the general population
32
prevention research
- health promotion
- universal prevention strategies
- selective prevention strategies
- indicated prevention strategies
33
health promotion
- looks at giving something to everybody to prevent later problems
- focuses on skills building rather than skills fixing, proactive vs preactive
- ie. teaching coping skills to people, parenting classes in schools
34
universal prevention strategies
- focuses on an entire population for health promotion to reduce a particular risk factor
- ie. tobacco use, alcohol use
35
selective prevention strategies
- targets an entire subgroup of a population who are at particular risk for something
- ie. children with single parents being at risk for various mental health outcomes
36
indicated prevention strategies
- targets those who have subclinical (early but not diagnosed) symptoms
- ie. teenagers with depressive symptoms targeted with certain prevention
37
genetics research
- family studies
- twin studies
- adoption studies
- twins reared apart studies
38
family studies
- examining symptoms in proband (the person with the disorder), first degree relatives, and more distant relatives
- if a disorder is more based on genes, then it should occur more often in close relatives than different relatives
- limitations: environment difference, people who are more related tend to live together
39
twin studies
- examining symptoms in MZ twins an same sex DZ twins
- if a disorder is based on genes, it should occur more in dizygotic twins
- advantage: any difference in mono/dizygotic twins must be due to genetics
40
adoption studies
- examining symptoms in adopted child, adoptive parents, and biological parent
- if the disorder is based on genes, the disorder should occur more in biological parents, and if the disorder is based on environment, the disorder occurs more in adoptive parents
- major advantage: genes and environment are completely seperated
41
twin reared apart studies
- direct test of heritability
- generally, as similar as twins reared together
- if twins reared apart show up with the same symptoms, we know that it is based on genetics
42
cross-cultural research
- interpreting incongruent findings from different cultures: how do you make sense of it? due to translation? need a culture specific theory?
- differences in symptoms: individuals may also present different symptoms, ie. tajin kyofoshu (a disorder in japan based on a fear of offending others)
- differences in cultural acceptability: depending on the culture and location, there's a difference in what presentations of symptoms are socially acceptable
- differences in treatment: in japan, psych workers take on parental roles vs medical model we have here
