PSYC - Ch 13 Flashcards
Types of Research
Qualitative
- Exploratory
- Small N
- Theory Building
- No hypothesis
- Narrative report
(non-numerical)
Quantitative
- Confirmatory
- Large N
- Theory testing
- Hypothesis
- Statistical Analyses
Quantitative Research
Category 1 - Descriptive
Category 2 - Correlational
Category 3 - Experimental
Quasi-Experimental
Non-experimental
Quantitative Research - Constraints
Level of constraints differ ie: how much control is exercised over the data
High constraint - more rules and more specific and precise methodology - higher precision & validity for
conclusions, BUT, not flexible
Low constraint - less rules and more specific and precise methodology - more flexible/exploratory but be careful about conclusions (precision? validity?)
Descriptive research (Goal and 3 Types)
Goal - to describe a single or multiple variables as they exist naturally
Not concerned with associations between variables
- Observational
- Survey
- Caste study
Observational Research
- Behavioral observation
- Content analysis
- Archival research
Behavioural Observation
- Direct observation
- Systematic recording or behaviours
- Typically in natural setting
Two concerns
1. Presence of researcher
- Solution = habituation
- Subjectivity
Solution = multiple raters, trained observers, clearly defined behaviors
Content Analysis
Indirect observation
Examines behaviours/events in literature, movies, television programs, or similar media.
ie: violence in cartoons
Archival Research
Indirect observation
Examines historical records to measure behaviors or events that occurred in the past.
ie: # of cases in which charges were dropped
Rules for Measurement of Descriptive Research
– Establish behavioural categories
– Employ multiple observers for at least part of the measurement process to obtain a measure of inter-rater reliability
– Select the technique that will be used to quantify the observations (frequency, duration or interval methods)
How to quantify observations
Frequency method
- Counting instances of behavior during
fixed-time period
Duration method
- Counting time spent engaging in behavior during fixed-time period
Interval method
- Dividing observation period into a series of intervals. Then, recording if behavior occurs during each interval. Finally, counting the number of intervals in which behavior occurred
Sampling Observations
Can’t observe everything all the time. Take samples of events.
Time sampling - observe in time frame, stop to record, observe again, stop again
Event sampling - observe specific events in specific intervals
Individual sampling - observe specific individuals in specific intervals
3 types of observation
Naturalistic / non-participant / unobtrusive observation
Participant observation
Contrived / structured observation
Naturalistic Observation
As if the researcher is not there - “fly on the wall”
Pros:
Behaviours observed in real world
High external validity due to authenticity
Best when behaviours cannot be manipulated
Cons:
Time consuming
Observer influence - curbed by habituation
Subjective interpretation - make a specific operational definition, clear/precise guidelines, multiple raters, inter rater reliability
ie: Jane Goodall when observing
Participant Observation
Interaction with participants
Only interacts with participants - OVERT
Becomes one of the participants without them knowing - COVERT (no informed consent here)
Pros:
Access to inaccessible info
Offers unique insight
High external validity due to authenticity
Cons:
Time consuming
Can be dangerous
Ethical issues (might have to do illegal things to keep your cover)
Observer influence (is Overt)
Loss of objectivity as researcher is in it
Contrived Observation
Structured observations - in a setting designed to illicit a behaviour
Pros:
Don’t have to wait for the behaviour to occur
Can be done in a lab or the feild
Con:
More artificial - observer is trying to trigger a behaviour
Less authentic
Lower external validity
Survey Research
Goal - get a sense of people’s attitudes and behaviours towards an issue
areas of consideration:
1. Question content/wording
2. Response format
3. Administering a survey
Question Content - Things to consider when asking a question
Be clear about the information needed
Avoid asking unnecessary questions but make sure to get the information you need
Best to ask demographic questions at the end so people remain honest throughout
Questions to aks yourself:
1. Is the Question Necessary/Useful? What level of detail? Maybe offer a range to be less intrusive.
- Are Several Questions Needed? Do you need further clarification?
- Is your Question Double-barrelled? Are you asking about multiple things? Be sure to ask about one thing only.
Don’t ask ie: What do you think of Concordia shuttle bus service and its drivers? - Do respondents have the needed information to answer the question? Do they know what you’re talking about?
Ask filtering questions beforehand
Is the terminology/acronyms too complex? - Is the Question Biased or Loaded? Your own blind-spots/bias may affect the wording.
Loads question - the way your question is worded, your participant
will have no choice but to admit to something they never engaged in.
“Have you stopped doing drugs?” What if you never did drugs? - Will Respondents Answer Truthfully? Some people are sensitive about answering certain things like age, income, taboo behaviours. - Use a range to curb this
Response Format
Open-ended Questions
Introduce a topic and allow participants to respond in their own words.
Ask the m to comment on a particular topic or issue.
ie: what are you views on X?
Give an indication of how much detail you want by the amount of space allowed and prompts
Restricted Questions
Present a limited number of response alternatives
ie: Multiple choice
You can add an element of open-endedness in a restricted question with “other (please specify)”
Rating-scale questions
Require selection of a numerical value on a predetermined scale
ie: On a scale from 1 to 10 state your interest in these topics
Likert-type scale - strongly disagree to strongly agree - 1 to 5 -
Assuming there is an interval even though we can’t know if they’re the same size.
Semantic Differential Scale - Multiple response options, polar opposites. Draw line across dash. - Reverse the questions to avoid a participant answering the same thing the whole way through
Administering a Survey
