Chapter 6 - Surveys and Observations Flashcards
define survey/poll
method of posing questions to people via telephone, personal interviews or written questinaires
define open-ended and forced-choice questions
open-ended: allow the responder to answer in any way that they would like too
forced-choice: the responder must show their opinion by picking the best of two or more options
define the likert scale
a survey question format using a rating scale containing multiple response options anchored by terms with strongly agree, disagree, etc.
define semantic differential format
a survey question format using a response scale whose numbers are anchored with contrasting adjectives. similiare to likert scale but just using different describing words rather than strongly agree.
define leading questions
these are problematic because its wording encourages one response more than the others, therefore tweaking its construct validity
define a double-barreled question
these are problematic because it asks two questions in one, weakening its construct validity
define negatively worded questions
these contain negatively phrased statements, making its wording complicated or confusing and potentially tweaking its construct validity
define response set
a shortcut respondents may use to answer items in a long survey, rather than responding to the content of each item.
also known as nondifferentiation
define acquiescence
answering yes or strongly agree to every item in a survey/interview. also known as yea-saying
define fence-sitting
playing it safe by answering in the middle of the scale for every question in a survey
define the faking good and faking bad phenomenons
faking good: giving answers on a survey that make one look better than one really is
faking bad: giving answers on a survey that make one look worse than one really is
define observer bias
a bias that occurs when observer expectations influence the interpretation of participants’ behaviours or outcomes of the study.
behaviour is rated according to their own expectations/hypothesis
define observer effects
a change in the behaviour of study participants in the direction of observers’ expectations. also known as the expectancy effect
define masked/blind design
a study in which the observers are unaware of the experimental conditions to which participants have been assigned
DOUBLE BLIND
define reactivity
a change in the behaviour of participants because they are aware they are being watched
define unobtrusive observation
made indirectly, through physical traces of behaviour, or made by someone who is hidden or is posing as a bystander
can observers be biases in observation research?
yes they may be.
to reduce:
- clearly operationalize variables to know how to observe
- have a clear coding scheme
- have multiple observers (inter-tester reliability)
can participants be biases in observation research?
yes they may be. they might just want to help out the research to confirm their hypothesis and just be a good participant
to reduce:
have blind study, use deception
what type of reliabilty is best used for observations
inter-rater
what are teh two things we look at too see if a measure is good?>
the reliability and validity (does it measure when they truly want to)
define construct validity
how well the researchers measured the variables
what are the different kinds of formats you can have for surveys?
open end: can provide more detailed data, and can help to explore a topic to learn about the common response. but people might not even asmwer the question. more about learning about topic
forced choice: is super easy to analyse and is more efficient. but it cannot fully describe how someone is feeling, its very general. more about quantifying data
what is important about the questions themselves?
wording!
BRUSO
Brief - not too long
Relevant
Unambiguous - not open to more interpretation
Specific
Objective
important elements about the wording of a question
- avoid leading question (to impose what the right response is)
- avoid baised language, be neutral
- ask multiple question sin different ways to get consistent data
- avoid double-barreled question (asking 2 in 1)
- avoud double negative (using negative words to be confusing)
- see if the order of questions has an impact or not (early question can influence response to later ones)
what are the ways in which people response to self-report measure?
non-differentiation: dont really read all the questions, just answer yes or no (reverse coding)
fence sitting: just play it safe and stay in the neutral positive the whole time (remove middle option)
social desirability: just wanting to look good (results stay annoymus)
some info is just hard to know about yourself: self-bias
relying on memories of past events: low accuracy for salient(most important) event, always reconstructing
what is the conscientiuos responder scale
this si when you put in question to make sure people are really reading.
choset the first option for this question
what are indirect measures of people submitting carelss responses?
response time: long time means that they are less focused
consistency indice: when does calculation, look at the consistency between questions of same construct