 Shuffle Toggle OnToggle Off
 Alphabetize Toggle OnToggle Off
 Front First Toggle OnToggle Off
 Both Sides Toggle OnToggle Off
 Read Toggle OnToggle Off
Reading...
How to study your flashcards.
Right/Left arrow keys: Navigate between flashcards.right arrow keyleft arrow key
Up/Down arrow keys: Flip the card between the front and back.down keyup key
H key: Show hint (3rd side).h key
A key: Read text to speech.a key
Play button
Play button
48 Cards in this Set
 Front
 Back
What does Operationalisation refer to?

Taking a meaningful but potentially illldefined concept, such as Age, for example, and transforming it into a precise measurement.


What three aspects of measurement is Operationalisation concerned with? (hint WHV)

What precisely is being measured.
How will it be measured. Values that the measurement can take need to be defined. 

With reference to study design, what is a Theoretical Construct?

The thing to be measured, such as Age, for example.


What is a Measure?

The tool to be used to make observations, such as a survey, for example.


Define a Variable.

The data corresponding to the observations made using a given measure.


Measurement?

The process of applying a measure to obtain data (the variable).


Nominal variable

An unordered collection of values where there is no particular relationship between values. Co our, for example.


Ordinal variable?

The set of possible outcomes have some meaningful ordering but the differences, ratios between any two values lack meaning. Arrival order, for example.


Interval variable

Natural ordering of values exists where differences may be meaningfully interpreted, but the ratios may not as no logical zero exists. Date of birth, for example.


Ratio variable

Natural ordering exists between values. Both differences and ratios may be meaningfully interpreted. Eg. Age in years


Discrete variables

Categoric, no guarantee that for any two values there exists a value that lies inbetween the two.


Continuous variable

Smoothly varying. For any two valid values, there is always a value that lies inbetween the two.


Non parametric variables

Nominal and ordinal scale variables


Parametric variables

Interval and ratio scale variables


Predictors

Variables used to predict the value of other variables. Also known as independent variables.


Outcome variables

Variables to be explained in terms of predictors. Also known as dependent variables.


Reliability

CPR
Consistency Precision Repeatability 

Validity

How accurate a measure is


Test retest reliability

Same measurement when retesting
Consistency over time 

Inter tater reliability

Consistency of measurement across people conducting a rating activity


Parallel forms reliability

Two different but equivalent mechanisms produce the same answer


Internal consistency reliability

Do equivalent sub parts of a measure yield the same answer.


Characterize experimental design?

Researcher exercises complete control over all relevant factors


Limitation of experimental design?

Can be too artificial


Non experimental design?

Some factors are left uncontrolled


Quasi experimental

A non experimental technique because some predictor variables are left uncontrolled.
Can analyze results. 

Case study.

A non experimental design where there is not enough data to analyze with formal statistical methods,


Internal validity

Does a study allow correct conclusions to be drawn regarding the causal relationships between predictor and outcome variables.


External validity

How generalizable are the results.


Construct validity

Does the measure appropriately capture the construct of interest.
Eg. Are you a racist? 

Face validity

Does the study look like it should work prima facie.


Ecological validity

Does the environment in which the study is conducted resemble the environment it seeks to explain in relevant ways.


What form of validity does a confound threaten?

A confound threatens internal validity.


Describe a confound.
What is it? Consequence? Where more likely? 
A confound is an unmeasured variable related to one or more predictors or outcomes.
It becomes difficult to understand the causal relationships between the predictors and outcomes. Non experimental research more likely to have confounds. 

What form of validity does an artifact threaten?

An artifact threatens external validity.


What is an artifact?
consequence? Where more likely? 
An artifact is where results apply in very limited circumstances.
Weakens genera usability of claims. More likely in experimental design 

History effect

Confound where specific events occurring mid study can affect the results


Maturational effect?

Confound where the change in a person (unrelated to a single specific event) during a study may affect the result.
Eg. Research on small children 

Repeated testing effect

Confound where results affected due to:
Practice Familiarity with environment Auxiliary effect of experiment, eg. Boredom 

Selection bias

Confound where the population of interest has been sampled in some way which affects the study outcome.


Differential attrition

Form of selection bias whereby people drop out of study in a way that affects results.


Heterogenous attrition

Drop out rate differs across experimental conditions


Non response bias

Confound relating to selection bias, where not everyone responds to a survey, or not all questions are answered, for example.


Regression to the mean?

A variation of selection bias. Occurs when selecting data based on extreme values of some measure.


Experimenter bias

Is the experimenter shaping the outcome of the experiment in some way. Clever Hans, for example.


Demand effects/reactivity?

An experimental artifact arising as a result of the fact that the participants knew that they were being studied. Studying worker productivity, for example.


Descriptive statistics

Used to describe the specific sample of data that you have


Inferential statistics

Using your data to make a claim about a broader population than the one from which your data was drawn.
