Week 5 Flashcards
The process of quantifying a set of observations of a phenomenon
Measurement
The working definition a researcher uses for a concept
Conceptual definition
The process of defining the exact way you plan to measure the concept
Operationalization
Any factor, trait, quality, or condition that can be measured
Variable
Those defined by their attributes or characteristics. These attributes or characteristics are categorized into two categories: exhaustive and mutually exclusive
Nominal level of measurement
The process of assigning a variable measured at the nominal level a numeric value
Coding
Those defined by their attributes or characteristics that are categorized into two: exhaustive and mutually exclusive; however, the numeric value assigned to them by the researcher can be logically ranked, but the intervals between categories may not be equal
Ordinal level of measurement
Also have to meet the criteria of being exhaustive and mutually exclusive. The numeric value assigned to the categories by the researcher can be logically ranked, which is similar to variables, measured on the ordinal level of measurement. Variables measured on this level, unlike variables measured on the ordinal level, have intervals between the categories that are equal.
Interval level of measurement
Have all the characteristics of variables measured at the interval level, in addition to a true zero value. In other words, the zero does mean the absence of a construct or whatever is being measured; zero value has real meaning.
Ratio level of measurement
Occurs when the data collected do not accurately reflects the construct we are attempting to measure.
Measurement error
Occurs when there is a pattern to the error
Systematic error
Providing the same response to all or most of the items on a measure
Response set
The tendency of a person to respond in a positive manner to a question regardless of the question being asked
Acquiescent response set
Occurs when the participants respond in a way that conveys a favorable impression of themselves or in a manner they believe will be more consistent with the researchers views
Social desirability bias
Occurs when one group of study participants is more likely to recall the information requested than other groups of participants
Recall bias
Occurs when study participants improve or change their behavior simply because they know they’re being studied
Hawthorne effect
Occurs when there is no consistent pattern of measurement error
Random error
A diminished ability to reject the null hypothesis
Bias towards the null
The degree to which a measure is consistent or dependable
Reliability
Defined as the degree of agreement between the raters
Inter-rater reliability
The degree of agreement between the observers
Inter-observer reliability
A statistic that is used to quantify the degree to which two raters or observers are in agreement
Cohen’s Kappa
Ranges from zero, which indicates no agreement between the raters, to one, which indicates perfect agreement between the raters
Fleiss’ kappa
Assesses the stability or consistency of a measure over time
Test-retest reliability
A coefficient that can be used to assess the test-retest reliability of two administrations of a measure
Pearson product-moment correlation
Used to assess the test-retest reliability, and those situations involving more than two administrations of a measure
Intraclass correlation coefficient
Used to determine the consistency of items on a scale
Split-half reliability
Achieved when the items on a scale are consistent with each other
Internal consistency
Used to assess the degree of internal consistency of the scales items
Cronbach’s alpha
The degree to which the measure actually measures what it was designed to measure
Validity
Based simply on whether the measure appears to be valid “on its face”
Face validity
Assessed by asking experts on the topic to evaluate whether the items on the measure cover the full range of the concept
Content validity
Based on how the measure compares to a criterion
Criterion validity
Indicates how well the measure corresponds with the criterion that is measured at the same time
Concurrent validity
A type of concurrent validity that assesses whether a measure can differentiate between groups known to differ on the variable being measured
Known groups validity
Assesses the ability of the measure to predict a criterion that occurs in the future
Predictive validity
An assessment of how well the concept of the variable was translated into a measure
Construct validity
A measure has__________ when it is correlated with measures that it should be related to, based on theory or empirical evidence
Convergent validity
A measure has__________ when it is not correlated with other measures to which it should not be related to
Divergent validity
Measures that have established reliability, and validity
Standardized measures
Established scores that are used to determine if someone meets or does not meet a specific criterion
Cut-off scores
The literal translation of the words
Denotative meaning
Captures the weight or intensity of the words
Connotative meaning
One person translates from the original language to the second language
One-way translation
One person translates from the original language into the second language, and another person, then translates back into the original language
Back translation
A process in which the original items are translated. After comparing the original and translated items, the original items may be revised, and then another translation is produced. This process continues until equivalent items are obtained in both languages.
Decentering
A measure where the respondents complete the question
Self administered measure/questionnaire
A measure where an interviewer reads the questions to the respondents
Interview schedule
Questions in which the respondents are asked to provide their own answers to
Open-ended questions
Questions in which the respondent is asked to select an option provided by the researcher
Closed-ended questions
Usually contain the word no or not, and can reduce the reliability of the measure
Negatively worded items
Usually contains more than one issue or topic yet allows for only one response
Double-barreled questions
Questions that contain an assumption
Loaded questions
Questions that encourage the respondents to respond in a particular way
Biased questions
The presentation of items on a survey, where the respondents are asked to write the responses using a specified scale
Likert scale format
Items on a scaled survey
Likert items
The sum of responses of the Likert items
Likert scale
Focuses on establishing equivalence by examining how items function across groups
Measurement equivalence
The extent to which a measure is compatible with the language and culture to which it will be adapted
Linguistic equivalence
The extent to which the concepts represented by the items making up the measure are understood to have the same meaning across cultural groups
Conceptual equivalence
Equivalence in item or question difficulty across groups
Metric equivalence