Association for Behavior Analysis International

The Association for Behavior Analysis International® (ABAI) is a nonprofit membership organization with the mission to contribute to the well-being of society by developing, enhancing, and supporting the growth and vitality of the science of behavior analysis through research, education, and practice.


Seventh International Conference; Merida, Mexico; 2013

Event Details

Previous Page


Paper Session #24
Methodological Discussions and Directions
Monday, October 7, 2013
12:30 PM–1:50 PM
Izamal (Fiesta Americana)
Area: EAB
Chair: Rebecca A Sharp (University of Auckland)
Using Calibration and Interobserver Agreement to Assess the Need for Training Novice Observers
Domain: Applied Behavior Analysis
KATRINA J. PHILLIPS (University of Auckland), Oliver C. Mudford (University of Auckland), Douglas Elliffe (University of Auckland)
Abstract: Visual inspection is the most common form of data analysis conducted within applied behavior analysis. For the analysis from visual inspection to be meaningful the collected data must reflect the actual target behavior. In order to increase the likelihood that the collected data do reflect the actual target behaviour, observer training prior to observers starting to make observations is recommended. However, observer training is seldom mentioned in research articles and there is little research to date on the topic. The current study used calibration and three interobserver assessments (block by block, exact agreement, and time window analysis) to assess the data collected by 4 novice observers recording a client's behavioral excess. Assessments were conducted before training, after training (where necessary), and during direct observation of the challenging behavior. The current presentation will discuss the utility of calibration and interobserver agreement and make recommendations for observer training.
How Representativeness is Affected by Dimensions of Behavior and Sampling Methods
Domain: Applied Behavior Analysis
REBECCA A SHARP (University of Auckland), Oliver C. Mudford (University of Auckland), Douglas Elliffe (University of Auckland)
Abstract: Rebecca Sharp, Oliver Mudford & Douglas Elliffe (The University of Auckland) Dimensions of behavior such as duration, rate and temporal distribution, have been demonstrated to affect the representativeness of observation samples. For example, Mudford, Beale and Singh (1990) found that when observation length was decreased relative to total time of interest, high-duration behaviours were more accurately recorded than low-duration behaviours. Similarly, the sampling method used can affect the representativeness of a sample. Although a number of studies have evaluated discontinuous sampling methods, few studies have evaluated the representativeness of samples obtained through continuous sampling. The current study simulated behavior, varying the parameters of each dimension of interest (i.e. rate and/or duration), and then simulated both continuous and discontinuous sampling of the simulated data. The effect of behavioral dimension and sampling methods will be presented, and the implications for practice discussed.

Effectively Translating Indirect Assessments with an Eye Toward Reliability and Validity

Domain: Applied Behavior Analysis
CRISTINA VEGA (SEEK Education, Inc.), Michele D. Wallace (California State University, Los Angeles), Marisela Alvarado (Private Practice), Adriana Gracias (A.B.E.D.I. Inc.)

Indirect functional behavior assessment instruments are used to preliminarily identify potential functions of behavior and have only been available in English. This study developed a Spanish version of the Questions About Behavioral Function (QABF) indirect assessment instrument. Forward and back-adaptation committees were used in the translation process. Subsequently, the QABF and the QABF-Spanish (QABF-S) were administered to a group of 80 bilingual participants to assess both the reliability and validity of the assessments. A factor analysis yield four factors that were consistent with the four subscales examined in this study. Reliability coefficients were good for the attention, escape, and tangible subscales on both the QABF and QABF-S, but were poor for the nonsocial subscale on each instrument. A second experiment was conducted to analyze the convergent validity between the QABF-S and functional analyses results with clients in the natural setting. Results, limitations and future research is discussed.

Low-Cost Portable Device for Collecting Behavioral Data Based on the Arduino Board
Domain: Applied Behavior Analysis
CARLOS ALEXIS PÉREZ HERRERA (National Autonomous University of Mexico), Rogelio Escobar (National Autonomous University of Mexico)
Abstract: In applied behavior analysis data are collected frequently in natural settings and in several occasions by the clients. Portability, accuracy and low-cost are important aspects in such recordings. Often, data has to be recorded manually in sheets of paper or by using custom-made wrist bands, thus accuracy is compromised. Several electronic devices are available for data recording but in most cases they are not portable enough for continuous use like laptop computers, or are expensive and not easily available like pocket PCs. In some cases, smartphone applications could be used but they can interfere with other tasks assigned by the users. We describe the design of a new dedicated device to collect behavioral data that consists of an Arduino board, a micro SD shield, and an LCD monitor. The device allows recording up to six events that can be stored in real time in the micro SD card for subsequent analysis. The events are recorded by means of push buttons and feedback is provided with the LCD monitor. This device has a compact design and it can be assembled easily, with no previous knowledge in electronics, and uses low-cost components that can be found easily around the world.



Back to Top
Modifed by Eddie Soh