Glossaries

• Attrition – The loss of participants during a study. For example, if an experiment started with 10 participants and 5 dropped out, the loss of the 5 participants is called attrition.
• Blind/Blinding – A practice where individuals (e.g., researchers, participants) are prevented from knowing certain information (e.g., goals/outcomes of an experiment, participant placement in experimental or control conditions) that may somehow influence their behavior. If both researchers and participants are blind, it is called a double-blind experiment. For example, an experiment involved testing the effects of a new drug. Some participants were placed in the group that received the drug and others in the placebo group in which they received a sugar pill. None of the participants knew which group they were in (i.e., they were blinded) because knowing could influence what symptoms they report.
• Confounding variable – An event during a study that the researcher was not studying, yet may have influenced the participant’s behavior, making the effects of the intervention less clear. For example, at the same time that a researcher introduces a language intervention at school, speech therapy (confounding variable) is started at home.
• External Validity – Likelihood that the same outcomes from a study would be observed if the intervention were applied outside of the study context (e.g., with different participants, in a different setting).
• Functional relation – A cause and effect relation. Experimental demonstrations that show that a reliable and consistent change in the behavior(s) addressed in the study is only produced by the intervention. For example, a teacher introduces a reading intervention with a student and reading improves. Reading does not improve for other students until the reading intervention is introduced with each of them. Here, there is a demonstration of a functional relation between the reading intervention (cause) and improvement in reading (effect).
• Internal Validity – Degree to which the outcomes observed in a study are solely due to the intervention and no other variables.
• Interobserver agreement – A procedure in which data from two or more independent observers is compared. High degrees of agreement between the observers increases the believability of the data.
• Parsimony – The practice of considering simple, logical explanations before considering more complex or abstract explanations.
• Procedural fidelity (Treatment integrity) – Data collected on the researchers’ implementation of the study procedures. These data are used to examine if the researchers implemented the procedures as intended.
• Pseudoscience – Theories and practices falsely claiming to be based on the scientific method.
• Randomization – A research method that involves assigning participants to groups at random. This process is used to ensure that the results of a study are not affected by bias.
• Reliability – The consistency between the events that actually occurred and the data that were recorded. For example, if a study included measuring instances of challenging behavior, we would have reliability if 10 instances of challenging behavior occurred in an hour and 10 instances of challenging behavior were reported for that hour.
• Replication – Repetition of all or any parts of an experiment to confirm previous findings.
• Social validity – A measure of the client, caregiver, and/or other relevant person’s perspective about the intervention. It includes some or all of the following:
  • the importance of the goals of the intervention
  • the acceptability of the procedures used in the intervention
  • the importance of the outcomes produced by the intervention
• Validity – The extent to which a study measures what it intends to measure. For example, if a study includes an intervention designed to decrease challenging behavior, a valid measure would be observing the number of instances of challenging behavior.

• Experimental design – A systematic approach to planning and conducting experiments to test if there is a relationship between variables.*See also alternating treatments design, changing criterion design, multiple baseline design, multiple probe design, randomized controlled trial (RCT), reversal/withdrawal

• Single-subject design – A systematic approach to planning and conducting experiments, that focuses on individual participants, to test if there is a relationship between variables. It has the following core features:
  • The behavior of the individual(s) before intervention is compared with their behavior after the intervention has been implemented.
  • The effect of the intervention is evaluated using visual analyses.
  • The behavior of the individual(s) is measured repeatedly across time/phases of the study.
    *See also – alternating treatments design, changing criterion design, multiple baseline design, multiple probe design, reversal/withdrawal

• Randomized Controlled Trial – A form of research in which participants are randomly assigned to two or more groups. One or more groups receive intervention and one group serves as the control. The control group may receive no intervention, a placebo, or continue to receive the intervention that was in place before the study started. The effectiveness of a specific intervention is determined by comparing the responses of the participants in the different groups.
  
  *See also – group design

• Group design – A systematic approach to planning and conducting experiments, that focuses on data collected from two or more groups of individuals, to determine the effectiveness of a procedure/treatment/intervention.
  *See also randomized controlled trial (RCT), quasi-experimental design

• Multiple baseline design – This is a type of single-subject experimental design. Procedures are replicated across different people, settings, or behaviors to see if there is a relationship between the intervention and the behavior targeted for change. The intervention is applied sequentially to each person, setting, or behavior to identify if the intervention is responsible for the change.

Graph retrieved from: Slocum, T. A., Joslyn, P. R., Nichols, B., & Pinkelman, S. E. (2022). Revisiting an analysis of threats to internal validity in multiple baseline designs. Perspectives on Behavior Science, 45(3), 681-694.

• Multiple probe design – This is a type of single-subject experimental design. Procedures are replicated across different people, settings, or behaviors to see if there is a relationship between the intervention and the behavior targeted for change. In this design, data are collected intermittently prior to introducing the intervention. The intervention is applied sequentially to each person, setting, or behavior to identify if the intervention is responsible for the change.

Graph retrieved from: Alqraini, F. (2017). Single-case experimental research: A methodology for establishing evidence-based practice in special education. International Journal of Special Education, 32(3), 551-566.

*See also single-subject design, multiple baseline design

• Reversal/Withdrawal – This is a type of single-subject experimental design. An intervention is implemented, then removed in an attempt to show that a desirable change in the target behavior(s) only occurs when the intervention is implemented. The target behavior reverses to undesirable levels when the intervention is removed. The intervention is implemented and removed multiple times to demonstrate that the change in the target behavior is due to the intervention.

Graph retrieved from: Fienup, D. (n.d.). PC – ABAB reversal graph. Instructional Design Lab. https://blogs.cuit.columbia.edu/df2675/graphing-tutorials/graphing-tutorial-basics/windows-pc-excel-graphing-tutorials/pc-abab-reversal-graph/

*See also single-subject design

• Alternating treatments – This is a type of single-subject experimental design in which two or more interventions are alternated (e.g., implemented on different days) in order to compare their effects on the target behavior(s).

Graph retrieved from: Kinugasa, T., Cerin, E., & Hooper, S. (2004). Single-subject research designs and data analyses for assessing elite athletes’ conditioning. Sports Medicine, 34(15), 1035-1050.

*See also – single-subject design

• Changing criterion – This is a type of single-subject experimental design. Goals for the target behavior(s) are systematically adjusted, in a stepwise fashion, to gradually change the behaviors from baseline levels to the desired terminal goal.

Graph retrieved from: Plavnick, J. B., & Ferreri, S. J. (2013). Single-case experimental designs in educational research: A methodology for causal analyses in teaching and learning. Educational Psychology Review, 25(4), 549-569.

*See also – Single-subject design

• Systematic literature review – A scholarly paper or a peer-reviewed article that uses a structured approach to comprehensively identify and synthesize existing research on a specific topic.
• Meta-analysis – A scholarly paper or a peer-reviewed article that uses a structured approach to comprehensively identify and synthesize existing research on a specific topic. A meta-analysis differs from a literature review because it uses statistical analyses to evaluate the data reported in the existing research.
• Quantitative research – A systematic approach to studying a topic through the collection and analysis of numerical data.
• Qualitative research – A systematic approach to studying a topic through the collection and analysis of descriptive data (e.g., natural language, participants’ expressions).
• Quasi-experimental design – A form of research in which groups of participants receive different types of intervention. The effectiveness of a specific intervention is determined by comparing data from the participants in the groups. In quasi-experimental design, participants are not randomly assigned to different groups (in contrast to randomized controlled trial design).*See also – group design, randomized controlled trial

• Baseline phase – A phase of a study in which data are collected on a participant’s performance before an intervention is introduced.
  *See also – Control condition
• Intervention phase – A phase of a study in which data are collected on a participant’s performance while the independent variable (e.g., teaching procedure, medication) is in effect.
  *See also – Experimental condition
• Maintenance phase – A phase of a study in which data are collected on the participant’s performance after a period of time has elapsed since the intervention phase concluded.
• Generalization phase – A phase of a study in which data are collected on a participant’s performance of a previously taught skill in an untaught location or form. This may include:
  • Performing the taught skill in a new context (e.g., settings, people, materials).
  • Performing an untaught skill that is similar to the taught skill.
• Control condition – A condition in an experiment in which the intervention is not present. For example, in a group design experiment, when testing a new program for teaching sight words. One group of students receives the new intervention. The other group of students receives standard instruction (this is the control condition). In a single-subject design experiment, when testing a new program for teaching sight words, a student would receive the new intervention for one set of words. The student would receive standard instruction for a different set of words (this is the control condition).
• Experimental condition – A condition or phase of an experiment in which the intervention is present. For example, when testing a new program for teaching sight words. One group of students receives the new intervention (this is the experimental condition). The other group of students receives standard instruction. In a single-subject design experiment, when testing a new program for teaching sight words, a student would receive the new intervention for one set of words (this is the experimental condition). The student would receive standard instruction for a different set of words.

*See also – intervention phase

• Duration – The amount of time during which a behavior occurs.
• Event recording – The recording of each occurrence of a behavior.
• Fluency – A measure of accuracy and speed of responding. An accurate response is fluent when it occurs quickly and without hesitation.
• Frequency – The number of times a behavior occurs.
• Latency – Amount of time between a cue (e.g., instruction, request, signal) and when the individual begins engaging in the behavior.
• Magnitude – The measure of amount, size, or intensity of a person’s action or an environmental event. For example, a low-intensity tantrum with crying versus a high-intensity tantrum with kicking and throwing, or the high versus low volume of the sound of a fire alarm.
• Momentary time sampling – A measure of whether the behavior is occurring at the moment that a specified time interval ends. The observer does not record data on the behavior if it occurs at any other time during the interval. For example, if a teacher is collecting data on on-task behavior using momentary time sampling. The teacher would set a timer for 1-minute. When the timer rings, the teacher would record if the student was on-task.
• Partial interval recording – A measure of whether the behavior occurred at any point during a specified time interval. For example, if a teacher is collecting data on on-task behavior using momentary time sampling. The teacher would set a timer for 1-minute. If the student is on-task at any time during the minute, the teacher records that they were on-task.
• Permanent product recording – A measure of behavior based on the presence of tangible evidence that the behavior occurred. For example, a written worksheet of completed math problems or an observation that a dirty room is now clean.
• Probe data collection – A system of measurement where data is taken intermittently (not on every response) during the data collection period. For example, data is taken on the first opportunity to tie shoes during the school day and is not collected the rest of the day (even if other shoe-tying opportunities occur).
• Rate – A measure of how often behavior occurs during a specific time period.
• Whole interval recording – A measure of whether the behavior occurred continuously during an entire specified time interval. For example, if a teacher is collecting data on on-task behavior using momentary time sampling. The teacher would set a timer for 1-minute. If the student is on-task for the entire 1-minute interval, the teacher records that they were on-task.

• Data Path – A line connecting data points that represents change over time.
• Data Point – A single point of data on a graph which represents a single measure of behavior.
• p-value – It is used to show whether an intervention is statistically significant. The p stands for probability and measures how likely it is that the observed change is due to the intervention. If the p-value is less than 0.05 then the results are considered to be statistically significant which means that the change was most likely caused by the intervention.

*See also – statistically significant

• Statistically significant – If something is statistically significant, it means that the results of the experiment are more likely to be caused by the intervention than simply due to chance. In most studies, a p-value of less than 0.05 would indicate that the results are statistically significant.

*See also – p-value

• Baseline data – Information collected on a participant’s performance before an intervention is introduced.
• Intervention data – Information collected on a participant’s performance while the independent variable (e.g., teaching procedure, medication) is in effect.
• Maintenance data – Information collected on the participant’s performance after a period of time has elapsed since the intervention.
• Generalization data – Information collected on a participant’s untaught performance of a skill. This may include:
  • The taught skill in a new context (e.g., settings, people, materials).
  • An untaught skill that is similar to the taught skill.

• Peer-reviewed article – A scientific article that has been objectively reviewed for quality and scientific rigor by experts and is edited, as needed, before publication.

• Assent – Assent is the active, voluntary, and ongoing permission provided by a participant or client, who is not legally authorized to provide informed consent for themselves, to participate in services or research. It can be expressed verbally (e.g., saying, signing, or using an AAC to indicate “yes”) or non-verbally (e.g., interacting with materials, walking over to the activity). For example, a participant under 18 years of age willingly agrees to participate in a study by saying yes.
• Bias – When a researcher’s actions are influenced, intentionally or unintentionally, by their values, beliefs, expectations, or desires about the study. For example, selecting participants who are most likely to be successful or scheduling the study procedures during regular work hours precluding working individuals from participating.
• Blinding – Withholding relevant information that could influence one’s judgment, decisions, or actions. For example, not telling a data collector whether or not a participant is receiving intervention.
• Confidentiality – Protecting a person’s identifying information, and any data entrusted to professionals during clinical (assessment, diagnosis, and intervention) and research participation.
• Conflict of interest – A situation in which an individual’s personal interests (e.g., family, friendships, financial, social) could compromise their professional judgment, decisions, or actions. For example, a clinician receives a financial incentive for referring clients to a specific service provider.
• Informed consent – Permission provided by a person, who has the legal right to consent, before participating in services or research, or allowing their information to be used or shared. For minors or individuals who lack decision-making capacity, permission is provided by their legal guardian. Before providing permission, the person must be informed of and understand the following:
  • Description of the procedures
  • Potential risks and benefits involved
  • That they are free to withdraw their permission at any time
  • Information about what to do if concerns arise