Chapter 2 Flashcards
A concept, for example temperature and weight
Construct
When a construct is not directly observable, observable measures representing the actual
construct are substituted
Indicators
The process of selecting an indicator(s) of a construct is called ___________
operatizing
How an indicator is defined for measurement is its ________
operational definition
Concepts such as “workplace violence” have several distinct subdimensions
or subconstucts are ________
multi-dimensional
characteristics or measures that can be several possible values (as opposed to constants which have only one possible value such as the speed of light sans gravity).
Variables
A variable’s set of possible values is called its
range
Variables that only have one value, for example weight, sex, body temp
Single-valued
variables that have multiple values, for example, race, majors
multi-valued
type of variables that can be any value in their range, for example, the temperature can be 70, 70.5, 81.6, depending on precise measurement
Continuous variable
type of variables that can only be specific values in a variable’s range, for example, school range, 1st - 12th grade (no 1.5 grades)
Discrete variables
Variables that can only be one of two possible values are often called indicator variables and can also be called
dichotomous or dummy variables
- the variable characteristic that has values that can be ranked and/or their differences calculated
- can be discrete or continuous
Quantitative
- variable characteristic that can only be categorized
* always discrete
Qualitative
common in academic research to represent abstract constructs that are not directly observable or measurable.
latent variable
how much quantitative information the variable provides for analysis.
level of measurement
Four levels of measurement in rank order
- nominal
- ordinal
- interval
- ration
- Level of measure that can only be classified
- discrete and qualitative
- usually single-valued
nominal variable
- Level of measurement of a variable that can be classified but can also be ranked, for example, a letter grade in a class
- always discrete and quantitative
ordinal variable
- Level of measurement of a variable that can be classified, ranked, the difference between values can be calculated, and the differences between values are consistent but it does not have a true, absolute zero
- the difference between values are always equal
Interval variable
• Level of measurement of a variable that includes interval variables but has a true, absolute, non-arbitrary zero.
ratio variable
The combination of indicators into a single measure for a construct
composite measure
the two types of composite measures often called measurement instruments
Index and scales
presume indicators are separate components that together form the concept, for example, the many components of job satisfaction
index
composite measure where the indicators are seen as causing the construct
formative constructs
a composite measure that presumes indicators are reflections of a concept
scales
composite measure where the construct is seen as causing the construct and the indicators reflect its variation
reflective constructs
The field that concerns the development and testing of composite measures in psychology
and education.
psychometrics
indicators that are simple yes/no type items denoting the existence of an attribute or event
indexes
Type statistics that provide summary information organizing and describing the data whether a sample or population
Descriptive
Type of statistics that use sample data to infer the value of an attribute or relationship in a full population.
Inferential
Descriptive and inferential statistics can be divided by the number of variables
they entail.
Univariate involves one, bivariate two, and multivariate statistics include three or more variables.
Most common bivariate descriptive technique which provides a crosstabulation of the frequencies for two variables simultaneously
“crosstabs”
contingency table
variables that are seen as affecting, or causing, others
independent variables
Variables that are being affected
dependent variables
Variables that are not affected by any others but are seen as causing one or more are
exogenous
Variables that have at least one variable affecting it are
endogenous
