Simultaneous testing of McNemar's problem for several populations

McNemar's problem concerns the hypothesis of equal probabilities for the unlike pairs of correlated binary variables. We consider four different extensions to this problem, each for testing simultaneous equality of proportions of unlike pairs inc independent populations of correlated binary variables, but each under different assumptions and/or additional hypotheses. For each extension both the likelihood ratio test and the goodness-of-fit chi-square test are given. Whenc=1, all cases reduce to McNemar's problem. Forc ≥ 2, however, the tests are quite different, depending on exactly how the hypothesis and alternatives of McNemar are extended. An example illustrates how widely the results may differ, depending on which extended framework is appropriate.     

SENSORY EXTINCTION AND SENSORY REINFORCEMENT PRINCIPLES FOR PROGRAMMING MULTIPLE ADAPTIVE BEHAVIOR CHANGE

The role of sensory reinforcement was examined in programming multiple treatment gains in self-stimulation and spontaneous play for developmentally disabled children. Two phases were planned. First, we attempted to identify reinforcers maintaining self-stimulation. Sensory Extinction procedures were implemented in which auditory, proprioceptive, or visual sensory consequences of self-stimulatory behavior were systematically removed and reintroduced in a reversal design. When self-stimulation was decreased or eliminated as a result of removing one of these sensory consequences, the functional sensory consequence was designated as a child's preferred sensory reinforcer. In Phase 2, we assessed whether children would play selectively with toys producing the preferred kind of sensory stimulation. The results showed the following. (1) Self-stimulatory behavior was found to be maintained by sensory reinforcement. When the sensory reinforcer was removed, self-stimulation extinguished. (2) The sensory reinforcers identified for self-stimulatory behavior also served as reinforcers for new, appropriate toy play. (3) The multiple treatment gains observed appeared to be relatively durable in the absence of external reinforcers for play or restraints on self-stimulation. These results illustrate one instance in which multiple behavior change may be programmed in a predictable, lawful fashion by using “natural communities of sensory reinforcement.”     

Returning a favor and retaliating harm: The effects of stated intentions and actual behavior

Two studies were carried out to determine the joint effects of an-other's stated intentions toward helping a person, and the amount of help the other actually gives a person, on the reciprocation of help. In Study I (n = 120) another person stated she intended to either help, not help, or did not state her intent; this person actually helped the subject a great deal or very little. The data showed subjects reciprocated solely on amount of actual help received (p < .001); however, questionnaire data revealed that when the past helper stated her intentions, regardless of whether the intent was to help or not to help, subjects showed less involvement with the situation. It was hypothesized that knowledge of the past helper's intent threatened the subject's freedom of choice with regard to reciprocation, which led subjects to ignore the intent in order to restore their freedom. Study II (n = 60) was a replication of Study I with the exception that the helper's intent, while stated, was discovered by subjects in such a way that they believed the helper unaware of the subject's knowledge. Both the other person's intent (p < .10) and her actual helping behavior (p < .001) were related to reciprocation. The results were interpreted in terms of Brehm's reactance theory, although other potential explanations were also explored.     

Redundancy analysis an alternative for canonical correlation analysis

A component method is presented maximizing Stewart and Love's redundancy index. Relationships with multiple correlation and principal component analysis are pointed out and a rotational procedure for obtaining bi-orthogonal variates is given. An elaborate example comparing canonical correlation analysis and redundancy analysis on artificial data is presented.A Fortran IV program for the method of redundancy analysis described in this paper can be obtained from the author upon request.     

The question-answer exchange between mothers and young children

Investigated associations between the questioning and answering behaviour of mothers and children in a design that allowed analyses within as well as between social classes. Thirty-two six-year-old children volunteered their knowledge and asked questions of their mothers about a variety of objects in a natural context. Social class differences in the children were found in the quality of questioning and the amount of knowledge displayed, but for the latter there were differences between tasks, and the incidence of questioning confounded the predictions made. Class differences in mothers' behaviour likewise differed by task. An index of the mothers' provision of cognitive meaning and of feedback was shown to correlate highly with children's rates of questioning and knowledge revealed both within and across classes.     

Classroom treatment of autistic children: II. individualized instruction in a group

This study was designed to assess the feasibility of providing individualized instruction in a classroom of autistic children. In particular, we investigated whether students' unsupervised responding was a variable which would influence the effectiveness of individualized instruction procedures. Initially, one teacher attempted to teach different academic behavior to four autistic children simultaneously. During these sessions the teacher rotated from child to child, providing individualized instructions and reinforcers to each. We found, however, that each child stopped responding whenever the teacher rotated to a different child and, further, no academic progress occurred for any of the children during these sessions. In a multiple baseline design, treatment procedures were then introduced (prompt fading, chaining, and programmed instructional materials) to teach each student to continue working for an extended period of time after each instruction from the teacher. After treatment, sessions of individualized instruction in a group were reintroduced for each child. The results showed that unsupervised responding was an important variable influencing the effectiveness of individualized instruction. Specifically, the data show that (1) unsupervised responding, acquired during treatment, generalized to the group setting; (2) after treatment, each child began to make progress on his assigned tasks; and (3) programming instructional materials in small steps appeared to be necessary in order to ensure the maintenance of unsupervised responding across changes in instructional materials. By the end of this study it was possible for one teacher to simultaneously teach four autistic children — with each child learning at his own rate, on his own task, with minimal supervision from the teacher. It is suggested that the generalizability of these results, to other settings where continuous supervision is difficult, may be an important concern of future research.     

SOME DETERMINANTS OF THE REINFORCING AND PUNISHING EFFECTS OF TIMEOUT1

Some determinants of the reinforcing and punishing properties of timeout were investigated in two experiments. Experiment I began as an attempt to reduce the frequency of tantrums in a 6-yr-old autistic girl by using timeout. Unexpectedly, the result was a substantial increase in the frequency of tantrums. Using a reversal design, subsequent manipulations showed that the opportunity to engage in self-stimulatory behavior during the timeout period was largely responsible for the increase in tantrums. Experiment II was initiated following the failure of timeout to reduce the spitting and self-injurious behavior of a 16-yr-old retarded boy. Using a multiple-baseline design, the nature of the timein environment was shown to be an important determinant of the effects of timeout. When the timein environment was “enriched”, timeout was effective as a punisher. A conception of timeout in terms of the relative reinforcing properties of timein and timeout and their clinical implications are discussed.     

RESEARCH ON THE DIFFERENCE BETWEEN GENERALIZATION AND MAINTENANCE IN EXTRA-THERAPY RESPONDING1

Many authors have reported that the development of programs for producing durable extra-therapy responding lags behind the development of programs for producing initial behavior change. In Experiment I, responding was recorded continuously in both the therapy and extra-therapy settings. The results showed that one child did not generalize to the extra-therapy setting, but that other children did. However, for the children who generalized, extra-therapy responding was not maintained. Therefore, in Experiment II two variables affecting the durability of extra-therapy responding were assessed and found to be influential: (a) the use of partial reinforcement schedules in the original treatment environment; and (b) the presence of noncontingent reinforcers in the extra-therapy environment. The results suggest that there are two distinct parameters of extra-therapy responding: generalization and maintenance. A technology for producing durable extra-therapy responding is discussed in terms of different treatment procedures required for different deficits in extra-therapy responding.     

Self-awareness and physical aggression

A theory of self-awareness was applied to aggression. In Experiment I, men were given an opportunity to aggress against women in the presence of a mirror, an audience, or neither. The mirror significantly inhibited aggression, but the audience did not. In Experiment II, an audience inhibited aggression, but only when there was frequent eye contact between the aggressor and the audience. These results were consistent with the predictions derived from the theory.     

The effect ofsome andall on reaction time for semantic decisions

Two experiments tested a model proposed by Meyer (1970) to account for the times required to verify semantic-memory statements quantified byall orsome. Each S was presented with bothall andsome statements in a mixed list, and the discriminability of false statements of the two quantifier types was controlled. In Experiment I positive subset statements ("horses are animals") were verified more quickly when quantified byall rather thansome; the reverse ordering occurred for negative subset statements ("horses are not animals"). Sentences with pseudowords in subject or predicate position took longer to reject than false real-word sentences. These findings contradict :Meyer's theoretical predictions and suggest that his earlier results were artifactual. Experiment II replicated the faster verification of positive subset statements quantified byall. This result was further shown to be predictable from the frequency with which Ss gave the predicate as a completion ofAll/Some S are _. The production frequency of predicates which form subset statements was lower when the quantifier wassome rather thanall. However, holding predicate production frequency constant, sentences with different quantifiers were verified equally quickly.