7 Module (Descriptive Methods)
16:45
Descriptive Research Designs Naturalistic Observation Designs Observe and describing behavior in a natural setting Not an experiment, so you can't show causation Start with genera problem statements Gradually narrow and focus as you learn more. But has greater ecological validity than most other research methods Can generalize to more real life situations How to observe: Undisguised observation Participants are aware that the researcher is observing their behavior 2 approaches: Nonparticipant observation you don't do what the thing you are observing is actually doing Participant observation you are actually getting involved with what your subject is doing (interacting with participants even just interviewing them like the love canal in New York) Disguised Observation Participants are unaware that they are being observed Options: Hide yourself (nonparticipant observation) with one way mirrors or being really far away Blend in not let the participants know that you are observing them (Rosenhan's Study) Use technology tracking with chips or collars
Expectancy Effects We selectively notice things we are looking for or expect. Other details slip past our perception. Naturalistic observation is especially prone to this, since data is often qualitative or subjective Solutions: Careful training Multiple observers Use video/audio so you can review Pros: Realistic Setting ( high ecological validity) Study must approximate real-life situation that is under investigation to have high ecological validity Gives you an idea of baseline behavior (before any manipulations) Good when you don't know much about the phenomenon Cons: Can't establish causality Not an experiment Not even a quasi-experiment (once you manipulate things, not naturalistic observation anymore) At best we can find some correlations, get a basic picture of what's relevant, see `normal behavior patterns' Often involves subject judgment. Vulnerable to expectancy effects. Concentrating on a small sample gives detail but may not be generalizable. Can be costly and time-consuming Laboratory Observation
Aka Systematic observation Observing behavior in a more contrived setting and focusing on a small number of carefully chosen participants Participants are more likely to know they're being observed So...the researcher can be: Participant or nonparticipant Disguised (altruistic research) or undisguised Pros: Usually costs less and time-consuming They are contrived, offering the researcher more control Cons: Expectancy effects still based on subjective judgments Greater possibility of subject reactivity Less flexibility Data collections in observational 3 designs ways to collect data: behavioral measures (direct and indirect) narrative records full narrative descriptions of a participant's behavior Considerations: Very subjective Can't be analyzed quantitatively Time consuming Checklists Researcher tallies attributes of the participants and whether certain behaviors were observed 2 types: Static checklist (attributes that don't change) Action checklist (things that do change) Considerations: qualifiable but... Reliability and validity are at risk May miss interesting behaviors
16:45 Descriptive Research Designs (Contd.)
16:45 Case Studies Narrative description of an individual or some aspect of an individual that brings together relevant aspects of the person's history and present situation Based on the logic of describing, analyzing, interpreting and evaluating a set of events and relationships within the framework or history. Data collected via in-depth interviews, documents, questionnaires, test results and archival records Also collect contextual and life history data They are also used to look at potential, not always negative behavior. Pros: Suggests hypotheses for future research Ability to present clinical implications of a particular disorder. Ability to describe processes not easily reduced to a single variable Cons: Participant may be atypical so you can't generalize Expectancy effects because it is a researcher observing. If based solely on self-report: honesty of participants can be off Being used now to study body integrity identity disorder where people want to have one or more body parts removed. Archival Methods Using data that existed before the study. Accomplished by obtaining data from: Organizations that collect data (school, government, etc.) Newspapers or magazines, libraries, or computer databases (also video or audio or drawings)
16:45 Descriptive in nature because the person compiling the data did not have any direct interaction with the data. Advantages: Reactivity is less of a problem, but is not completely removed Less time-consuming Disadvantages: Selection bias what information was included in the database Conclusions are based on data collected by someone else how can we know for sure whether the data are reliable or valid? Qualitative Methods Focuses on phenomena that occur in a natural setting. always takes place in the field (aka field studies) Data are analyzed without statistics Researcher is interested in interpreting and making sense of what they have observed Data collection: Observation and/or unstructured interviewing in natural settings. Spontaneous and open-ended You can't analyze it and what about subject reactivity?