Lecture 3: Research Methods in Relationship Science Flashcards
why talk about research methods?
Foster critical thinking and media literacy
limits of personal experience & intuition
- Confirmation bias
- We aren’t objective observers of our relationship interactions
- Our perspective is unique and limited
- We’re too quick to see causal relationships where there may be none
confirmation bias
the tendency to seek out, pay attention to, and believe the evidence that supports our preexisting beliefs
the scientific approach
Relationship scientists examine questions systematically by collecting data and testing their hypotheses while taking steps to reduce bias, and subjectivity, and rule out alternative explanations
can we study love?
- Studying love is possible, but challenging
- Things research scientists are interested in, like love, commitment, satisfaction, trust, and attraction, are all very abstract
- How we operationalize these psychological constructs is key
operationalizing our constructs
- We can’t measure these psychological constructs directly
- We need to carefully consider whether we are really measuring what we think we are measuring
measurement validity
the extent to which an operationalization adequately captures the psychological construct of interested
measurements should
- Be well-grounded in existing theoretical conceptions of the construct
- Relate to other measures of the construct & predict future outcomes
different types of measurement
Self-report
Behavioural observations
Physiological measures
self-report
Simply ask participants to describe their thoughts, feelings, and behaviours
two types of self-reports
fixed-response questionnaire & open-ended question`
fixed-response questionnaire
a specific set of questions and possible responses are pre-determined by the researcher (ex. The love scale)
open-ended question
- The participant gives any answer that comes to mind
- Helpful when studying something that we don’t know much about yet
- A way of gathering information to generate more specific questions later on
qualitative research
a methodological approach relying primarily on open-ended questions
content analysis
examine the broader themes that emerge from participants’ responses
pros of self-reports
- Inexpensive
- Easy to administer
- No special equipment is required
- Allows us to recruit more participants
- Allows us to “get inside people’s heads”
cons of self-reports
- Difficulty with self-awareness and recall
- Social desirability bias
- Participants may not interpret questions in the way you intended
social desirability bias
desire to be seen in a positive light
sentiment override
global beliefs about the partner/relationship may colour perception and memory of specific interactions
example of sentiment override
“how many times did your partner kiss you yesterday?”
“I don’t remember, but he loves me and we have a good relationship, so it must have been a lot.”
behavioural observation
- Gather data about relationship events without having to ask people who are experiencing those events directly
- Train observers to watch & code recordings of participant behaviour
- We can observe people anywhere (with participants’ consent)
- We need to decide which behaviours to observe
- Coders must agree on what constitutes incidents of a given behaviour category
- A lot of behaviours require interpretation
- Requires extensive training
interrater reliability
the extent to which coders agree on whether a specific behaviour has or has not occurred
pros of behavioural observations
- Directly assess behaviours of interest
- Don’t have to rely on faulty memories
- Avoid social desirability bias
cons of behavioural observations
- Expensive
- Time-consuming
- Labour-intensive
- Reactivity
reactivity
a change in one’s behaviour caused by the knowledge that one is being observed
indirect measures
- Designed to avoid reactivity and social desirability
- Ex. reaction time can be used to assess implicit attitudes
implicit relational attitudes test
- Instructions: indicate the valence of the word as quickly as possible
- Index of automatic attitudes = negative word -RT - positive word RT
- A higher score = a more positive attitude
underlying premise of implicit relational attitudes
some concepts are more tightly linked in our minds based on experience
implicit vs. explicit attitudes and relationship satisfaction
- Conscious/explicit attitudes are not correlated with implicit attitudes
- Implicit but not explicit attitudes are associated with newlywed’s changes in satisfaction over time (4 years)
- Implicit attitudes predict nonverbal behaviour in couple’s discussions
- Nonverbal behaviours are linked to satisfaction with conversation and relationship satisfaction over the following week
- Explicit attitudes did not predict either verbal or nonverbal behaviour
pros of indirect measures
- Avoid social desirability bias & reactivity problems
- Could be particularly useful for sensitive topics
cons of indirect measures
- Big gap between the construct of interest and operationalization
physiological responses
- the body’s reaction to various experiences/stimuli
- Ex. autonomic nervous system, hormone changes, brain activity, and immune system changes
challenges collecting and interpreting fMRI data
- Very confined & noisy environment
- Makes it challenging to create powerful psychological experiences for participants
- Don’t see brain activation per se: we infer activation by subtracting responses on the control trial from the trial of interest
- Need to think carefully about task design
pros of physiological measures
- Interesting in their own right (ex. Understanding the link between relationships and health)
- Outside participants’ control (not susceptible to social desirability bias
cons of physiological measures
- Very expensive -> smaller sample size
- Ambiguity in interpretation
- Can be more invasive (depending on the measure)
archival data
- Data that has been collected by others, often for other purposes
- We might be able to draw on publicly available data & documents
- Ex. more positive facial expressions in yearbook photos predict the likelihood of being happily married 30 years late
pros of archival data
Typically economical
Can examine historical trends
cons of archival data
Limited by the type and quality of the original data
takeaways: measurement
- No single approach is perfect & free of limitations
- Ideally, we want to adopt a multimethod approach: using a combination of methods to triangulate an answer
correlational design
Allow us to examine naturally occurring associations between variables
interpreting correlational data
- The strength of association is captured by the correlation coefficient (r), which can range from -1 to +1
- The sign tells us the direction
- The magnitude tells us the strength of the association
pros of the correlational design
- Sometimes the only option available
- We can’t manipulate some variables
cons of the correlational design
We can’t conclude causation
does marriage cause happiness?
- There is evidence to suggest that marriage & happiness are associated
- It’s plausible that marriage may increase happiness
- But, it could also be true that happier people are more likely to get married
- Or, there may be a third variable responsible for the effect
factors necessary for causation
- Two variables must be correlated
- One variable must precede the other
- There must be no alternative explanation for the association
cross-sectional data
data collected at one single point in time
longitudinal data
- Data collected from the same participants on multiple occasions
- Allows us to examine change over time
challenges and considerations of longitudinal research
- choosing the right interval
- the sample you start with may not be the same one you end up with
daily diary study
a type of longitudinal approach where participants provide data every day at about the same time
experience sampling
a type of longitudinal design where data is gathered throughout the day, thereby capturing behaviours, thoughts, and feelings as they occur
attrition bias
the participants who drop out may systematically differ from those remaining in the study
pros of longitudinal research
- Captures change over time
- Can examine processes that would be impossible/unethical to cause
- Daily diaries and experience sampling are less subject to retrospective bias
cons of longitudinal research
- Expensive
- Time and labour-intensive
- Attrition bias
- Getting one step closer to making causal claims, but still not there
statistically controlling alternative explanations
- You measure an alternative variable that might explain your effect, and you include it in your analysis
- If your effect is just a proxy for something the control variable is doing your effect will be wiped out
- Although helpful, this type of statistical control is still subject to limitations
- It can be hard to anticipate every relevant variable
experimental design
- Manipulate one variable to determine the effect on another variable
- The only design that allows us to make causal claims
- Compare the experimental group to the control group to determine the effect of experimental intervention
- We want the control condition to match the experimental condition as closely as possible (minus the IV) to avoid confounds
random assignment
- Every participant has an equal chance of being assigned to an experimental or control group
- With a large enough sample, our two groups should be similar in all these individual traits
- This helps rule out alternative explanations
internal validity
- Can we rule out alternative explanations in the experiment?
- Relies on the selection of appropriate control & random assignment
longitudinal experimental studies
- Could expose couples to an intervention and track them over time (longitudinal assessments)
- The same concerns about attrition bias apply
pros of experiments
Allow us to make causal claims as long as there are no threats to internal validity
cons of experiments
- May have lower external validity: the extent to which results obtained in a given study generalize to other contexts
- Not always an option
population
all the people we are interested in
sample
a subset of that population
choosing a sample
Must be chosen carefully to ensure it represents the population we want to generalize to
can we make universal claims about relationships?
- We often make universal claims about relationships, but we don’t study diverse samples, so we can’t always make these claims
- Ex. homosexual couples are historically underrepresented in relationship research
convenience samples
- Anyone who is readily available
- Easier to get, but may not be representative of people in the broader population
- They are used more frequently
WEIRD participants
Western, Educated, Industrialized, Rich, and Democratic
representative sample
- A sample that resembles the entire population we want to study on the variables of interest
- Very difficult to get
volunteer bias
the people who agree to participate may differ from those who don’t
who participates in couples research study
- Researchers mailed invitations to participate in a longitudinal study to couples who had obtained their marriage license in LA County between 1993 and 1994
- Those who responded to the invitation
1) Higher SES (more years of education, higher status job)
2) More likely to have cohabitation prior to marriage
what differentiates couples who participate from those who don’t?
- Individuals whose partner agreed to participate with them reported greater relationship satisfaction, commitment, and more secure attachment
- Couples who agreed to participate together were less likely to experience a breakup overtime
actor-partner interdependence model
Allows us to examine how individual outcomes are affected by both one’s characteristics (actor effect) and the partner’s characteristics (partner effect)
ethical issues
- Asked to deeply think about and confide about issues of a highly personal and sensitive nature
- They may experience negative effects, like recognizing problems in the relationship for the first time
- We need to weigh the cost of doing the studies, the cost of not doing them, and the benefit of doing them
- We need to be sensitive to how we approach our participants, provide effective debriefing, and counselling resources
