A note on analyzing ordered alternatives

A general method of devising stepwise multiple testing procedures with fixed experimentwise error is applied to the problem of non-parametric randomized block design with ordered alternatives. In addition, the method is applied to other models with ordered alternatives.The author wishes to thank the referees for helpful suggestions.     

Do the mentally retarded have poor memory?

The hypothesis that mentally retarded persons have a short-term memory deficit whixh is due to inadequate rehearsal strategies is reexamined. A reevaluation of results of a number of experiments suggests that normal and retarded persons differ on short-term memory tasks, and the superiority of normals is present from the time of initial stimulus exposure. While some evidence for slope differences in retention curves is found in a few tasks, it is not strong support for memory differences. The hypothesis that “memory” differences are due to differential encoding as a result of more adequate rehearsal by the normal subjects is unacceptable. If this were the case, some differences in slope of retention curves could be expected unless rehearsal beyond the encoding stage was totally precluded. The review points to a need for an empirical reevaluation of the issue with a full recognition that deficits in particular processes may characterize some retarded persons and not others.     

Cognitive changes associated with social skills training

Changes in locus of control and self-esteem during social skills training and at 6 month follow-up were investigated. Forty-four adolescent, male offenders were randomly assigned to a social skills training, attention-placebo, or no-treatment control procedure. The results showed that pre- to post-training increases in self-esteem were produced for both the social skills training and attention-placebo control groups, but not for the no-treatment control group. These increases did not continue during the follow-up phase; the social skills training group, which showed the greatest improvement during training, displayed a decrease in self- esteem during follow-up.The locus of control measures showed a statistically significant shift towards internality for the social skills training group during training, which was not found for the attention-placebo and no-treatment controls. During the follow-up phase, all three groups showed a statistically significant shift towards externality, suggesting a tendency for boys to view their behaviour and consequences as being under the control of chance factors or powerful others, during institutionalization. This effect was partly reversed during training for the social skills training group, but the effect was not long lasting. It seems therefore that social skills training is effective in producing an increase in self-esteem and a shift towards internal locus of control. These changes are, however, of short duration and the self-esteem shift may be the result of increased staff contact or other non-specific therapy factors.     

Treatment of bisexual pedophilia by a biofeedback-assisted self-control procedure

A bisexual pedophile was treated for 88 days using a biofeedback procedure which displayed his erection response to sexually deviant stimuli on a closed-circuit television monitor. Told to develop a strategy of self-control using the biofeedback, his response to young boys and girls was treated sequentially in a multiple baseline design. His concurrent response to boys, girls and adult males was evaluated in separate generalization sessions. The treatment was successful in suppressing deviant response while maintaining nondeviant response in strength. Debriefing of the client following treatment revealed that he had used a self-developed technique of covert sensitization to suppress deviant response and had used the biofeedback to confirm and validate this self-control strategy.     

The effects of a response cost treatment tactic on hyperactive children

In a pilot study, a multielement design was used to assess the effectiveness of a response cost procedure on a 7-year-old child's hyperactive behavior and academic performance across days. The procedure was effective in reducing off-task behavior and in increasing academic assignment completion. In Experiment 2, three strategies were compared to a no-treatment baseline in treating an 8-year-old hyperactive child: drug (Ritalin) alone, response cost alone, and drug plus response cost. The cost program alone and the cost program combined with medication were effective in reducing off-task behavior and in increasing academic performance. In both studies, the procedure was viewed by the teachers as practical and effective for use in a classroom setting. The subjects liked the cost system and believed that they completed more academic work when it was operative.