Chapter 4 - Operationalization Flashcards
Research Details and Design
What is operationalization?
The development of specific definitions for all relevant factors related to the research topic of interest.
Operationalization turns vague or ambiguous concepts into detailed descriptions, which can be measured.
In order to operationalize, a researcher needs to identify indicators logically related to the concept of interest.
What is an independent variable and dependent variable?
The independent variable is the one presumed to influence or cause the dependent variable.
What is a mediating variable?
Sometimes called the intervening variable is something that influences the mechanism by which the IV affects the DV
A mediating variable may be a factor in WHY the IV-DV relationship exists in the first place.
Ex: If we think our intervention increases fruit and vegetable consumption by first increasing awareness of places to purchase food, this awareness would be considered a mediating variable.
What is a moderating variable?
Moderating variables influence the strength or direction of the IV-DV relationship, and contextualize the effect between the IV and DV
What is controlled variable?
Variables in research that are accounted for in different ways.
Controlling for the effects of variables is a common in public health research studies.
Researcher holds constant balances across groups, or neutralizes so it will not have an effect on the study.
What is constant variable?
A constant variable has no variation. It is the same for all cases in a study, and helps control the effects of the variable.
What is a positive correlation?
In a positive correlation, the value of the DV increases or decreases in the same direction as the IV. They both go up or they both go down.
What is a negative correlation?
In a negative correlation, the variables change in opposite directions. As one goes up, the other goes down.
What is a curvlinear correlation?
A curvlinear correlation occur when the relationship between the variables transforms at certain variable levels.
What is a conceptual framework?
Outlines the theoretical or logical relationships among variables, and the progression from the IV to the DV.
The concepts and associated variables are organized and linked sequentially.
These frameworks are typically based on theory, evidence, or specific knowledge. Developing a conceptual framework is useful for considering proposed causal linkages between the variables in the study and other key factors potentially influencing the outcome (e.g. moderating variables(
When developing a conceptual framework, start with the end point and work backwards to identify the potential relationships among the variables.
What is a hypotheses?
Hypotheses are statements that predict the nature and extent of relationships between variables based on your research question or research goals in quantitative studies.
What is a null hypothesis?
The null hypothesis states that there no differences in attributes, effect, or relation between variables.
What is an alternative hypothesis?
The alternative hypothesis affirms the existence of differences in attributes, effect, or relationship between variables.
What is a good hypothesis?
Is based on your research question Predicts the outcome of your research Is directly related to the variables of interest Is clear and concise Is testable
What is a dichotomous or binary variable?
Variables with only two attributes (completing intervention or not completing intervention)
What are polychotomous variables?
Variables that have more than two attributes. They can be categorical or continuous.
What is are the four levels of measurement?
- Nominal - categories with no numerical value
- Ordinal - scale in which each value is rank ordered in a way that is higher or lower than the previous value, but intervals are unknown or not equal
- Interval - Scale in which values have order and equal intervals, without a true zero point
- Ratio - scale in which values have order and equal intervals, with a true zero point.
What is systematic error ?
Systematic error reflects an imprecise measure due to personal views or beliefs, and it affects the way people answer questions there by distorting true responses.
A common reason for systematic error due to bias is social desirability.
What is social desirability?
Social desirability is when people tend to answer questions in a way that they think is socially acceptable.
Social desirability can result in acquiescent response set bias
What is acquiescent response set bias?
Agreeing or disagreeing with all statements in order to provide what one thinks to be favorable responses.
What is random error?
Responses may overestimate or underestimate true values in unpredictable ways , resulting in random error.
They are due to chance and have no consistent pattern.
Sometimes factors such as mood or circumstances influence responses.
What are two ways to evaluate the quality of measurement procedures?
Reliability
Validity
What is reliability?
The degree of consistency something has. The more reliable a measure is , the less likely it is to be impacted by random error.
What types of test are there for reliability?
Test-retest reliability - measure of stability over time. Administering an instrument to collect data from a sample of people and having those same people complete the same measure again.
Internal consistency reliability - items throughout a measure should be consistent, whereby answers to questions relating to similar concepts should be highly associated with one another.
Inter-rater reliability - degree of agreement or consistency among multiple observers.
What is validity?
Validity refers to the extent to which a measure is a true reflection of a real meaning of the concept it is intended to measure.
What are the different types of validity?
Face Validity - subjective judgement of whether a measure is a reasonable way to measure some concept or if a logical relationship between the variable and the proposed measure exist.
Content Validity - the degree to which a measure covers an entire concept.
Criterion Validity - means scores on one measure are similar to scores on another measure of the same condition or criterion
Predictive Validity - occurs when scores on one measure can predict the scores on a similar measure.
Concurrent validity - a type of criterion validity that exist when scores on one measure are closely related to scores on another measure
What is “r” in a linear association?
The sample correlation coefficient (r) is a measure of the closeness of association of the points in a scatter plot to a linear regression line based on those points, as in the example above for accumulated saving over time.
What is internal validity?
The extent to which we can infer that the independent variable caused the dependent variable.
What is external validity?
The extent to which the findings will generalize to other populations, settings, measures, and treatments
What is measurement validity?
The quality of accuracy of individual measures or scores. The extent to which a score measures what it was intended to measure.
How do we assess internal validity?
Equivalence of Groups on Participant Characteristics
Low - Groups very different, marked differential attrition
Medium - Some attempts to equate groups or groups found to be similar
High - Random assignments to groups and low attrition
Control of Experiences and Environment Variables (Contamination)
Low - Extraneous variables not controlled, no comparison group
Medium - Attempts to control experiences and environment
High - All extraneous variables controlled, eliminated, or balanced.
How do we assess external validity?
Population
Low - actual sample unrepresentative of the theoretical population
Medium - some attempts to obtain good sample
High - actual sample representative of theoretical population
Ecological (naturalness of setting/conditions, extent to which results are restricted to a specific time in history)
Low - Unnatural setting, tester, procedures, and time
Medium - somewhat artificial (Questionnaire)
High - Natural setting, tester, procedures and time
What are best practices for operationalization?
Look to past studies for operationalization and measures
Include members of the population of study in defining/modeling
Reduce chance for measurement error
Use existing valid and reliable measures if possible
Keep track of the process of operationalization and measure choice
