Making systematic observations and asking questions Flashcards
Behavioural Measure
recording actual behaviour of subjects
- frequency
- latency (time taken to respond)
- number of errors
Physiological measure
a participants bodily functions
self-report measures
common form: rating scale
–> likert scale
(ordinal measure)
Implicit measures
measure unconcious responses
Reliability of a measure
the abiity to produce similar results when repeated
–> more variability - less reliability
margin of error
the likely variation from samle to samle
interrater-reliability
the degree of agreement between multiple observers
test-retest reliability
administering the same test twice, separated by a long interval of time
Parallel-forms reliability
form of test used on first administration, replaced on second administration by parallel form
Split-half reliability
two paralell forms in a single test
–> they are separated and scored individually
Accuracy of a measure
measure that produces results that agree with a known standard
Validity of a measure
the extend to which a measure measures what you intend to measure
Face validity
how well a measurment instrument appears to measure what it is designed to measure
Content validity
how adequately an instrumet/test measures a theoretical construct
Criterion-related validity
how adequately a test score can be used to predict an individual´s value on some criterion measure
- concurrent validity - if scores on a test are collected at the same time
- predictive validity - comparing scores on a test with the value of criterion measure observed at a later time
construct validity
e.g. is a test constructed in a way that it successfully tests what it is supposed to measure
range effects
occur when values of a variable have a upper or lower limit
- ceiling effect, a lot of cases have a very high score
- floor effect, a lot of cases have a very low score
role attitude cues
signal to a participant that a change in attitude is needed to conform to his/her new role as research participants
- cooperative attitude - strong desire to please experimenter
- apprehensive attitude - worrying what will happen
- negative attitude - participant tries to ruinn experiment
experimenter bias
the behaviour of the experimenter influences the results of the experiment
Single-blind technique
the experimenter doesn´t know which experimental condition a subject has been assigned to
- to reduce experimenter bias
double-blind technique
neither experimenter nor participants know which treatments the participants are receiving
- to reduce experimenter bias
Correlational (non-experimental) research
observing variables in nature, without manipulation
experimental research
manipulating variables
quantitative data
data yeilded by conting and quantifying behaviour; numerical
qualitative data
written records of observed behaviour; analyze qualitatively
frequency method
record the number of times a particular behaviour occurs
duration method
how long a particular behaviour lasts
intervals method
divide observation period into discrete time intervals, record whether behavior occurs within each interval
Ethnography
researcher becomes immersed in behavioural or social system being studied
participant observation
–> ethnography
acting as functioning member of a group (covertly)
nonparticipant observation
–> ethnography
observer as nonmember of the group (ovetly)
Sociometry
identifying and measuring interpersonal relationships within a group
Case History
descriptive technique
–> observation on a single case
(no experimental design)
Archival research
involves studying existing records
–> purely descriptive
Content analysis
Use when you want to analyze written or spoken record for occurrence of specific categories or events, items, or behavior
–> only yields qualitative data
qustionnaire items: Open-Ended Items
allow participants to respond in their own words
–> restricted items, but provided with an additional “other” category
questionnaire items: Restricted Items
provide a limited number of specific response alternatives
–> controls participant’s range of responses
questionnaire items: partially open-ended items
restricted items, but provided with an additional “other” category
–> gives opportunity to give an answer not listed
questionnaire items: rating scales
like likert scale
–> provides a graded response
non-response bias
large proportion of participants fail to complete and return the questionnaire