Levels of Measurement
Only gives a name/category, no ordering implied. Example: the numbers on the back of soccer jerseys
What is Nominal?
Assessing how practical the scale is to use in the field. Aspects to consider are the time it takes to administer, how easy to administer, language used to ensure the phrasing is clear/understood.
What is Utility?
They mirror 3 basic single case designs in traditional research.
What is non-experimental, quasi-experimental, and one type of true experimental?
What does SSD stand for?
Single-subject design
- Meaning (such as defined in a dictionary)
- A type of definition that tells you what your constructs are
- Abstract
-Example: Depression
What is Conceptual Definition?
Can be rank-ordered. Distances between attributes do not have any meaning. The interval between values is not interpretable in an this type of measure.
What is Ordinal?
Some constructs are relatively easy to assess while others require more subtle or indirect measurement.
Direct (weight, blood pressure, etc.) or indirect (concepts, such as anxiety).
What is Directness?
- Most frequently use single phase design
- Only has an intervention
- Allows for measurement of change over course of an intervention
- Non-experimental design (similar to 'One Group Post-Test Only' group design)
What is B - intervention only design?
What does the 'A' typically stand for in SSDs?
What is:
No intervention / baseline?
- Only tells you how to measure it
- How you can relate your studied constructs to a particular domain under your study
- Example: heart rate, anxiety rating scale ('on a scale from 1 to 10...')
What is Operational Definition?
In this type of measurement the distance between attributes does have meaning. The interval between values is interpretable. For this reason, it makes sense to have an average. Example: distance between 40 degrees and 90 degrees
What is Interval?
Concerned with "error in measurement." Example: how consistently or dependably does a measurement scale measure what it is supposed to be measuring? When conducting tests there will always be a degree of random error in the administration of measurement scales -this is the extent to which a score is free of random error. Can be equated with stability, consistency and dependability. Can be assessed multiple ways: test-retest reliability for stability, inter-item reliability for internal consistency and inter-rater reliability or parallel scale for equivalence.
What is Reliability?
- Has 2 phases which are *no-intervention baseline phase* AND *intervention phase*
- Quasi-experimental design
- Having a baseline allows for establishment of a relationship between intervention and outcomes
What is AB - Baseline and Intervention Design?
What does the 'B' typically stand for in SSDs?
What is:
Intervention?
Name the 4 types of measures commonly used.
What are: Standardized measures, observational measures, individual rating scales (IRS), and client logs?
In this type of measurement there is always an absolute zero and it is meaningful. You can construct a meaningful fraction (or *****) with a ***** variable. Example: 4 is a ***** variable. You can "count" variables - for example, the number of clients in past 6 months. You can have 0 clients and because it is meaningful to say that "...there were twice as many clients in the past 6 months as compared to the previous six months."
What is Ratio?
Assessing to establish whether or not it is reliably measuring what you want it to measure. Concerned with the meaning/interpretation of a scale. There are many ways of testing it. It is not absolute. It is a matter of degree rather than an "all or nothing concept."
What is Validity?
- Has 3 phases: *no-intervention baseline phase* then *intervention phase* and finally *no-intervention withdrawal phase*
- Allows for evaluation of pre-intervention, intervention, and post-intervention problem status
- More reliable type of design
- Causality CAN be determined
What is ABA - Basic Withdrawal Design?
Name 5 types of SSD (single subject designs).
What are: B-intervention only, AB - baseline and intervention, ABA - basic withdrawal, BAB - intervention-baseline-intervention, and Multiple Baseline Designs?
- A measure of internal consistency
- How closely related a set of items are as a group
- Tip:
.80 has relatively high internal consistency
.70 or higher is considered "acceptable”
What is:
Cronbach's Alpa / Coefficient Alpha?
The hierarchy of measurement levels, starting from the weakest to the strongest.
What is...
Nominal - only names attributes (weakest)
Ordinal - attributes can be ordered
Interval - distance is meaningful
Ratio - absolute zero
?
The difference between what exists in reality and what is measured by the research instrument/tool.
Can be caused by bias...
What is Measurement Error?
- When you may not be able to wait to establish a baseline with a client (ie. crisis situation) and must begin intervention immediately
- Intervention is followed by a baseline which is followed by the intervention.
What is BAB Design? (Baseline-Intervention-Baseline)
- Allows for evaluation across clients, situations, or problems
- Causality CAN be determined
- A true experimental design
- Useful in evaluating cases in which an intervention would likely result in enduring changes in the dependent variables
What are Multiple Baseline Designs?
What are some methodologies for describing interventions?
What are: narratives, videotapes, audiotapes, and/or new computer applications/programs (ie. Zoom)?