|
| Further Refinements of Observation and Measurement Procedures |
| Saturday, May 27, 2006 |
| 1:00 PM–2:20 PM |
| Regency VI |
| Area: DDA; Domain: Applied Research |
| Chair: Eileen M. Roscoe (New England Center for Children) |
| CE Instructor: Eileen M. Roscoe, Ph.D. |
| Abstract: This symposium will present four papers describing various refinements in observation and measurement procedures in the field of applied behavior analysis. The first paper, presented by Jeffrey Luke, will present data comparing whole-session means and latency to first response as methods for identifying the function(s) of severe problem behavior. The second paper, delivered by Maeve Meanny, will discuss data comparing partial-interval recording (PIR) and momentary-time sampling (MTS) methods for estimating duration and frequency. She will also present data evaluating how PIR and MTS methods may affect treatment interpretation. The third paper, delivered by Frank Symons, will review the strengths and weaknesses of recent developments in sequential analysis, focusing on issues relevant to contingency analysis in natural environments. Finally, the last paper, presented by Luanne Witherup, will discuss the utility of obtaining behavioral measures of running away for children in foster care. |
| |
| Utilizing Latency-to-First-Response as a Measure in the Evaluation of Functional Analysis Outcomes. |
| JEFFREY R. LUKE (University of Iowa), Joel Eric Ringdahl (University of Iowa), Tory J. Christensen (University of Iowa), Jayme Mews (University of Iowa), Jason M. Stricker (University of Iowa), Terry S. Falcomata (University of Iowa) |
| Abstract: Since the initial application of functional analysis logic to the assessment of severe problem behavior, variations in the methodology have emerged. Typically, such variations have included a change in design or in length of the assessment. In this study, we compared the graphic display of whole-session means and latency to first response as methods for identifying the function(s) of severe problem behavior. Visual inspection of the graphic displays latency to first response resulted in similar conclusions as visual inspection of whole-session mean data. These results will be discussed relative to their implications for session length during analog functional analyses and when such methodology should be employed (e.g., dangerous behavior, etc.). Inter-rater agreement was obtained for at least 30% of the comparisons and was above 90%. |
| |
| A Comparison of Momentary Time Sampling and Partial Interval Recording Measurement Methods. |
| MAEVE G. MEANY-DABOUL (New England Center for Children), Eileen M. Roscoe (New England Center for Children), William H. Ahearn (New England Center for Children), Jason C. Bourret (New England Center for Children) |
| Abstract: The purpose of this study was to compare partial interval recording (PIR) and momentary time sampling (MTS) methods for estimating duration and frequency, and to evaluate how PIR and MTS methods may affect treatment interpretation. Five individuals with autism, who exhibited problem behavior characterized by different bout durations, participated. Responding was recorded across baseline and treatment conditions using each type of measurement method to determine whether the different methods affected the data record, and thus, data interpretation. Results indicated that MTS was a better estimator of duration than PIR across all bout durations, and that PIR was a better estimator of frequency than MTS across short and medium bout durations. Long bout durations resulted in decreased accuracy of estimates when compared to frequency. Separate graphic displays were then created for each measurement method. A dual-criterion (DC) method was used to analyze each graph for evidence of appropriate phase changes and treatment effects. Visual inspection using the DC method resulted in appropriate phase-change decisions but varied treatment interpretations across measurement methods and bout durations. IOA was collected during 33.8% of sessions and averaged 95%. |
| |
| Calculating Contingencies in Natural Environments. |
| FRANK J. SYMONS (University of Minnesota), Jennifer J. McComas (University of Minnesota), Ellie C. Hartman (University of Minnesota), John D Hoch (University of Minnesota) |
| Abstract: In this paper, we rejoin the discussion initiated by Vollmer and Hackenberg (2001) on identifying and estimating reinforcement contingencies in the natural environment. The conceptual and logical basis for inferring a reinforcement effect is revisited. Recent developments in sequential analysis are reviewed along with their strengths and weaknesses. Data from descriptive analyses are used to illustrate points of convergence and divergence. Remaining issues germane to contingency analysis in natural environments are discussed. It is concluded that the conceptual distinctions among contiguity, contingency, and dependency are critical if the logic of sequential analysis is to be extended successfully to a behavior analytic account of reinforcement in natural environments. |
| |
| Baseline Measurement of Running Away Among Youth in Foster Care. |
| LUANNE WITHERUP (University of Florida), Timothy R. Vollmer (University of Florida) |
| Abstract: The purpose of this study was to examine the feasibility and usefulness of obtaining behavioral measures of running away for children in foster care. Participants included 84 runaways residing in one service district of the Florida Department of Children and Families (FDCF). All data were obtained from existing databases managed by FDCF. Seven baseline measures were calculated for each runner including (a) the number of run initiations, (b) the proportion of opportunity days in which the child initiated a run, (c) the number of days the child spent on the run, (d) the proportion of opportunity days that the child spent on the run, (e) the duration of successive run episodes, (f) successive episode inter-response times and (g) successive initiation inter-response times. The first four of these measures were also calculated for various sub-groups of children constructed from the original sample. In addition, an observer-panel evaluation was conducted to identify baselines that would be suitable for research-based treatment evaluations. Results demonstrate the feasibility of obtaining baseline measures of running away, but suggest that such measures may not be suitable for research-based treatment evaluations in many cases. However, results indicate that such treatment evaluations may be possible via an analysis of groups of runners rather than individual subjects. |
|
| |