Lecture 3: Approaches to Measurement Flashcards
Before observing behavior researchers need to address a few major questions what are these questions ?
will observations be conducted in a naturalistic (confounding variables are unknown) or contrived (an environment specifically created for research) setting?
will participants know they are being observed?
will they know why they are being observed?
how will behaviors be recorded? (temporal measures, checklists, narrative records, rating scales)
What are the three threats to the reliability and validity of observations? and how can they be minimized/reduced
Observer Bias : observers rate behaviors by own expectations rather than what they see (see what they expect to see)
Observer Effects : perceived expectations of the observer inadvertently change behaviors of participants to match the observers expectations
combat:clear operational definitions, multiple observers, training observers well, establishing inter-rater reliability
Reactivity : change in behavior because they know they are being watched
ex: clever hans
Combat: blending in or waiting it out (longer present the less concerned with their presence) to be less noticeable; one way mirrors however ethics
Interviews can be _________, ____________, or _________
structured : questions asked bever deviate from the list of questions
semi structured : conversation with purpose’ have a set of questions they want to ask but follow the interviewee in terms of phrasing, order, and structure of questions
unstructured : general topic in mind but no fix set of questions - goes wherever it goes
What are ways to improve the quality of responses obtained during interviews?
whats a downside?
Participants should be asked the same set of questions
questions should be open ended
all leading questions should be avoided if possible
it is important for interviewers to demonstrate active listening throughout the interview (verbal and nonverbal communications “wow” “interesting” “aha” etc.
HOWEVER, personal reactions should be avoided specifically approval or disapproval
*interviews are very time consuming so researchers prefer surveys and questionaires
surveys and questionnaires: Single item measures rely on what?
the analysis of one item
surveys and questionnaires: A multi-item scale looks at what?
the responses to a set of items to measure a single variable
What are the three different response formats for surveys and questionnaires?
Does wording matter ?
open ended questions
forced choice questions :
MCQ
likert scale
yes its important to prioritize neutral wording and ordering of questions
What is the social desirability response bias? how does it impact surveys?
its the desire to select what is believed to be the social desirable answer to look better in the eyes of others
it decreases a surveys construct validity (whether or not test measures what its supposed to)
*unsure if responses are truly what the participant believes or what they think they should be believing
what is acquiescence? one way to combat it? how does it impact survey validity?
selecting the same response throughout the entire survey ex: yes or all strongly agree
inserting reverse wording questions
or neutral question phrasing to avoid desirability in responses
weakens a surveys construct validity?
