Ledford & Gast - Chapter 1

Applied research

-Involves systematic investigation related to the pursuit of knowledge in practical realms or to solve real-world problems

-E.g., seek to identify interventions that result in improved physical activity for children with Down syndrome

Basic research

-Concerned with the advancement of knowledge that may or may not have immediate and specific application to practical concerns

-E.g., might inform science related to the association of running and behavioral abnormalities in a mouse model of Down syndrome

Independent variables

-Variables manipulated by the researchers

-The interventions

Dependent variables

-Variables we expect to change given the manipulation of the independent variables

-The target behavior(s)

The importance of incorporating scientific methodology into the daily routines of practitioners in schools, clinics, and communities

1) Advancement of science

2) Advancement of practice

3) Empirical verification of behavior change

Internal validity

-The extent to which an experiment shows convincingly that changes in behavior are a function of the independent variable and not the result of uncontrolled or unknown variables (Cooper)

-A study with adequate mechanisms for ensuring that outcomes are related to the intervention procedures rather than extraneous factors

External validity

-The degree to which a study's findings have generality to other subjects, settings, and/or behaviors (Cooper)

-

Experimental control

-Two meanings:

(a) the outcome of an experiment that demonstrates convincingly a functional relation, meaning that experimental control is achieved when a predictable change in behavior (the dependent variable) can be reliably produced by manipulating a specific aspect of the environment (the independent variable)

(b) the extent to which a researcher maintains precise control of the independent variable by presenting it, withdrawing it, and/ or varying its value, and also by eliminating or holding constant all confounding and extraneous variables (Cooper)

-Shows that the experimental procedures (intervention) and only the experimental procedures are responsible for behavior change

Functional relation

-A verbal statement summarizing the results of an exper-iment (or group of related experiments) that describes the occurrence of the phenomena under study as a function of the operation of one or more specified and controlled variables in the experiment in which a specific change in one event (the dependent variable) can be produced by manipulating another event (the independent variable), and that the change in the dependent variable was unlikely the result of other factors (confounding variables); in behavior analysis expressed as b = f (x1), (x2),c, where b is the behavior and x1, x2, etc., are environmental variables of which the behavior is a function (Cooper)

-When experimental control is demonstrated, a functional relation has been verified between the independent and dependent variables

Evidence-based practice

-Intervention procedures that have been scientifically verified as being effective for changing a specific behavior of interest, under given conditions, and for particular participants

Reliability

-Refers to the consistency of measurement, specifically, the extent to which repeated measurement of the same event yields the same values (Cooper)

-Ensure reliability by defining the target behavior/event clearly and concisely so that two independent observers consistently agree on scoring what they observe

Threats to internal validity

-Variables other than the planned independent variable that could result in changes in the dependent variable

Nomothetic vs idiographic

-Nomothetic research approaches are generally based in the natural sciences and are characterized by attempting to explain associations that can be generalized to a group given certain characteristics

-Idiographic research approaches attempt to specify associations that vary based on certain characteristics or contingencies present for the participant or case of interest

-Argued that an idiographic approach may be more appropriate in practice, especially in special education

Baseline logic

-A term sometimes used to refer to the experimental reasoning inherent in single-subject experimental designs; entails three elements: prediction, verification, and replication (Cooper)

-Study participants serve as their own control; used in single case designs , which are quantitative experimental research approaches

Validity

-The extent to which data obtained from measurement are directly relevant to the target behavior of interest and to the reason(s) for measuring it (Cooper)

-Accuracy

History

-Events that occur during an experiment, but that are not related to planned procedural changes, that may influence the outcome

Maturation

-Changes in behavior due to the passage of time

Testing

-Threat in any study that requires participants to respond to the same test repeatedly, especially during baseline or probe condition

-It it the likelihood that the repeated assessment task will result in participant behavior change

Instrumentation threats

-Refers to concerns with the measurement system

-Of particular concern with single case design because of repeated measurement by human observers who make errors

Procedural infidelity

-Refers to the lack of adherence to condition protocols by study implementers

Attrition

-The loss of participants during a study

-Can limit generality of results, especially if participants with certain characteristics are likely to drop out

Attrition bias

-The likelihood that attrition impacts the outcome of the study

Sampling bias

-Occurs in single case designs, when researchers use additional, non-explicated reasons for including or excluding potential participants

Data instability

-The amount of variability in the data (dependent variable) over time

Cyclical variability

-A specific type of data instability that refers to a repeated and predictable pattern int he data series over time

Multiple-treatment interference

-Can occur when a study participant's behavior is influenced by more than one planned "treatments" during the course of a study

-An interactive effect may be identified due to sequential confounding or carryover effect; minimize with counterbalancing

Regression to the mean

-The likelihood that following an outlying data point, data are likely to revert back to levels closer to the average value

Adaptation

-A period of time at that start of an investigation in which participants' recorded behavior may differ from their natural behavior due to the novel conditions under which data are being collected

Hawthorne effect

-Participants' observed behavior not being representative of their natural behavior as a result of their knowledge that they are participants in an experiment

Similarities between research and practice

Applied researchers and teachers/therapists must

(a) Be able to identify and analyze problems

(b) Generate creative solutions

(c) Implement an intervention in a systematic manner

(d) Document the effect of the intervention

(e) Act on the data in an ethical and responsible way

What are some threats to internal validity?

-History

-Maturation

-Testing

-Instrumentation

-Procedural infidelity

-Selection bias

-Attrition

-Attrition bias

-Sampling bias

-Multiple-treatment interference

-Cyclical variability

-Regression to the mean

-Adaptation

-Hawthorne effect

What are some ways to make Hawthorne effect less likely?

-Familiarize participants with experimental conditions

What are some ways to make adaptation less likely?

-Be as unobtrusive as possible

-Expose participants to unfamiliar settings, adults, formats, data collection procedures, etc. beforehand

What are some ways to make regression to the mean less likely?

-Continue collecting data until stability is established, as data may improve just based on variability of data and on the intervention itself

What are some ways to make cyclical variability less likely?

-Vary conditions lengths across time as this decreases the predictability and make it less likely that data collection will align with an unidentified natural source that could account for variability

What are some ways to make data instability less likely?

-Maintain the condition until there is stability or attempt to isolate the source of the variability

-Be patient and analytical in research decisions

What are some ways to make multiple-treatment interference less likely?

-Counterbalance experimental conditions

-Carryover effect can be seen with a visual analysis and can be minimized by collecting data until there is stability

What are some ways to make selection bias (attrition, attrition bias, and sampling bias) less likely?

-Include more participants than required in case a participant drops out

-Randomly select participants from a pool of participants who meet inclusion criteria

What are some ways to make history less likely to impact intervention results?

-Longer studies generally have a greater threat of history

-Take careful research notes to help explain the variability due to transient history effects

What are some ways to make maturation less likely to impact intervention results?

-Studies of shorter duration (4-6 weeks) are less likely to have an impact

-Session fatigue may be related to maturation and one should consider the participant's age and attention span and adjusting accordingly

What are some ways to make instrumentation less likely to impact intervention results?

-Attend to the details of your measurement system