Teaching Children How and When to Make ... - Science Direct

BEHAVIOR THERAPY 11, 509--521 (1980)

Teaching Children How and When to Make Emergency Telephone Calls RUSSELL T . JONES

Universtty of Pittsburgh ALAN E. KAZDIN

Pennsylvama State University The present study examined a behavioral procedure which was designed to teach young children how and when to make emergency phone calls. Experiment 1 compared the effectiveness of three conditions---behavioral training, teacherdevised training and no training--m teaching children how to make emergency calls. The conditmns were administered to six classes in two different preschool programs. The behavioral training program resulted in significantly greater improvements in emergency dialing skills relative to teacher-devised and no-training conditions. Experiment 2 examined a discrimination trailfing procedure with selected subjects from Experiment l to ensure that the children not only knew how to make the emergency phone calls but also knew when to make them, I.e., in what situations. In a multiple basehne design across children, the discrimination training procedure was shown to improve performance in identifying when to apply the emergency diahng skills.

Each year, 12,000 people die in fires, another 300,000 are injured as a result of fires, and 1,000,000 buildings are destroyed by fire in the United States alone (National Fire Protection Association, 1975). As a result, the National Fire Protection Association and the Bell Telephone Company have recently published booklets and instructional materials aimed at reducing losses due to fire (National Fire Protection Association; Project Telepac, 1976; Telezonia, 1975). Children constitute an especially Portt,ns of this paper were presented at the Fifth Annual Convention of the Association of Behavior Analysts, Dearborn, MI, June 1979. The authors would like to thank those students who served as raters or experimenters m various phases of this project. Special thanks are extended to Janell Haney for her competent assistance in the coordination of several aspects of this study. This project was supported by a grant (S07 RR07084-13) awarded to the first author by the Bmmedlcal Research Support Grant Program, National Institute of Health. Requests for reprints should be sent to the first author, Department of Psychology, Clinical Psychology Center, 703 Old Engineering Hall, Umverslty of Pittsburgh, Pittsburgh, PA 15260. 509 0005-7894/80/0509-052151 00/0 Copyright 1980 by Association for Advancement of Behavior Therapy All rights of reproducuon m any form reserved

510

JONES AND KAZDIN

important target population for this instruction because they were regarded as a cause of fires in 66,760 cases from 1970-74 (National Fire Protection Association) and because fire is the leading source of child fatalities in the home (Burger King fire safety program, 1979). Occasionally, investigations have reported training children to use the telephone (e.g., Left, 1974; 1975), but programs directed specifically at emergency telephone dialing have rarely been reported (Jones, 1980). Fire and telephone officials, teachers, and parents may assume that adequate emergency dialing skills will be obtained without specific training procedures or that adults will be available to make the emergency call. Emergency dialing skills do not, however, appear to be learned inadvertently (Jones), and adults are not always present or able to make the call. Also, even though children often report that they know when to use newly learned emergency phone skills, objective assessment does not support their claims (Jones). The purpose of the present study was to develop and to examine the effectiveness of a behavioral program designed to teach young children how and when to make emergency phone calls.

EXPERIMENT 1 Method Subjects and Settings An emergency dialing checklist (see Assessment) was administered to 65 children who were enrolled in either of two preschool programs in Western Pennsylvania. The programs were selected for participation on the basis of interest. Five children were excluded because of frequent absences or change of programs, and one was excluded because of possession of requisite behaviors on the pretest (score above 4). Thus, the subjects were 33 boys and 27 girls (mean age = 5 years, 1 month; range = 3 years to 6 years) from three classrooms in each program (classes having been selected on the basis of teacher interest), for a total of six classrooms? Training was conducted in the classroom by the teachers.

Assessment Dialing checklist.

The emergency dialing checklist (see Table 1) was used to assess each child's ability to call the operator and provide relevant information. Responses 10 through 14 were individually scored as "correct" or "incorrect." Responses 1 through 9 were scored in the same manner, but these responses had to be in sequence to be correct, i.e.,

i Subjects were obtained from two preschool programs" a H e a d Start p r o g r a m and a private program. Three c l a s s r o o m s from each p r o g r a m were c h o s e n to participate, and each was randomly assigned to one of three experimental groups. C l a s s r o o m s were selected from two different b u d d m g s within each program. B e c a u s e kindergarten-aged children were included in the private program, several 6-year-old c h d d r e n were participants in the p r e s e n t study.

511

EMERGENCY DIALING SKILLS TABLE 1 CHECKLIST OF DIALING SKILLS

Dtals 1. Picks up the recetver. (Push Dial Tone.) 2. Places receiver on ear. 3. Dials zero ( " 0 " ) .

Yes Yes Yes

No No No

Yes

No

Yes Yes

No No

Yes Yes Yes

No No No

Yes

No

Yes Yes Yes Yes

No No No No

" ' L e t ' s pretend that I am the o p e r a t o r . " " O p e r a t o r . "

Reports Fire 4. There is a fire at m y house.

Gives Name 5. First n a m e . 6. L a s t name.

Gwes Address 7. H o u s e n u m b e r . 8. Street name. 9. City Tell the child to put the receiver down. Tell the child to pick up the receiver and put it to his ear. " L e t ' s m a k e believe the operator is l i s t e m n g . " "Tell the operator what your full n a m e i s . " (TIME) 10. My n a m e is

(first, last)

"Tell the operator to send the firetruck to your h o u s e . " 11. 12. 13. 14

Send the firetruck to m y house. House number. Street name. City.

" P u t the receiver d o w n . " "Thank you."

when the child was marked "incorrect" on any response, all responses which followed it were automatically "incorrect." Scores could range from 0 to 14 (since each "incorrect" response counted as one point). The checklist was administered both before and after training, with a period of approximately 4 weeks between assessments. For each testing session, subjects were randomly assigned to and assessed individually by one of four undergraduate psychology majors who had received nine 45miD training sessions over a 2-week period. Testing, which required l0 to 15 miD, took place in a small room in the building of each school, with the experimenter and the child seated at opposite sides of a desk. Content validation. This checklist was adapted and modified from that of a previous investigation (see Jones, 1980) in which its content

512

JONES AND KAZDIN

was validated by (a) contacting city fire officials, who offered steps in making emergency calls, (b) observing a 4-year-old child making emergency phone calls and developing a checklist from these observations, and (c) obtaining feedback from two consultants from the Bell Telephone Company and the National Fire Protection Association, whose function was to provide children in classroom settings with information and with the materials to learn emergency skills, as well as from two specialists in child development and three parents with preschool and primary grade children. The second checklist was administered to two pilot subjects, ages 4 and 5, to ensure appropriateness of the checklist skills for the target population. Reliability of observation. All testing sessions were audiotaped, and reliability ratings were carded out by one undergraduate and one graduate student, who served as independent raters. Reliability was computed by dividing agreements of observations (appropriate or inappropriate responses) between observers by agreements plus disagreements and multiplying by 100 to form a percentage. Reliability, obtained across all 60 children, attained a mean of 98%. In addition to the audiotape recordings, direct observation of 10 (16.7%) of the subjects by a second observer led to a mean reliability coefficient of 93%, as computed with the above formula.

Experimental Design and Conditions A pretest-posttest control group design (Campbell & Stanley, 1966) was employed to evaluate treatment. Three classes within each of the two preschool programs were randomly assigned to either behavioral training, teacher-devised training or no-training control conditions. The behavioral training condition consisted of modeling, practice, feedback, prompts, rehearsal, and tangible reinforcement. Because emergency dialing skills could possibly be taught by merely instructing teachers to train the requisite emergency dialing skills and to conduct training independently of a specified behavioral training regimen, the behavioral training method was compared to a teacher-devised method. A no-training group was included in the design to control for changes resulting from history and maturation and from repeated assessment of the target skills. Behavioral training. This method included daily use of instructions, modeling, prompting, remediation, feedback, review, and reinforcement. Teachers were supplied with a teletrainer, a simulated telephone apparatus that allows one person to serve as an operator while another person calls. Training was carried out sequentially on the following tasks: dialing the operator, reciting one's name and address, and providing emergency information. Each of these tasks was divided into smaller components to facilitate training. Thus, dialing the operator was taught in the following steps: (a) picking up the receiver, (b) picking up and replacing the receiver, and (c) picking up the receiver, dialing " 0 , " listening for the operator, and replacing the receiver. Reciting name and address was

EMERGENCY DIALING SKILLS

513

also divided into component parts: (a) giving full name and (b) giving full address. Providing emergency information was first taught singly and was then required in combination with giving name and address. Instruction proceeded by teaching each of the three individual tasks and then combining them in sequence: (a) dialing the operator, (b) giving name and address, (c) providing emergency information, (d) providing emergency information, name, and address, and (e) dialing the operator and providing emergency information, name, and address. During lessons both the teacher and students played the role of a parent giving instructions to contact emergency agencies (the firetruck, the police, and the ambulance). Emergency situations included such stimuli as yelling "fire" and pretending to be disabled by an accident. Training was conducted in a group, but children practiced individually. Reinforcers .(M & M's and raisins) were provided for correct completion of the small component steps within each task, at first, and, later, for the entire sequence of desired behaviors. Teacher-devised method. Teachers in this condition were also given access to telephone equipment, including both the teletrainer and real telephones and were instructed to teach the children how to respond to emergency situations by calling the operator and how to provide the relevant information. They were encouraged to conduct training in any way they wished as long as they focused on the relevant skills mentioned above, but specific behavioral techniques were not encouraged or discouraged. Biweekly checks were made to ensure that training sessions were being held. No-training control. These teachers were told that their students would be assessed with respect to telephone skills during the school year. A 4-week period elapsed between pre- and posttesting, as with the other groups, but without intervening training experiences.

Results Preliminary Analyses Preliminary analyses of variance of pretreatment performance of the emergency dialing skills revealed no significant differences as a function of treatment conditions or subject gender, but students from the different preschool programs differed slightly in their ability to perform the dialing tasks, F(1,58) = 4.43, p < .05. Although subjects from one preschool program had a higher mean level of performance than did subjects at the other preschool program, M = 2.7 vs. M = 1.8, mean performance of both groups indicated the absence of the requisite skills prior to treatment. Because subjects from each school were included in each treatment condition, different performance levels associated with the programs would not differentially bias treatment effects. At posttreatment, girls tended to perform better than boys (p < . 10). However, subject gender did not interact with treatment conditions so that the major results, discussed below, were not differentially applicable to girls and boys.

514

JONES AND KAZDIN

Treatment Effects The effects of training on emergency dialing skills were evaluated with an analysis of covariance that compared the three conditions, using pretreatment performance as the covariate. Significant group differences were obtained, F(2,56) = 34.27, p < .001. The adjusted means for each of the groups, plotted in Fig. 1, were then compared using the Tukey's Wholly Significant Different Test, a more conservative multiple comparison technique recommended for unequal sample sizes (Games & Howell, 1976). These multiple comparisons revealed that behavioral training led to significantly higher levels of emergency dialing skills than did the teacher-devised training and no-training conditions (p < .05). Although the teacher-devised training method was superior to no training at all, this difference did not attain significance. In addition, within-group t tests from pre- to posttreatment demonstrated that the behavioral training group [t(17) = 7.43, p < .001] and the teacher-devised training group [t(22) = 3.62, p < .01] improved in emergency dialing skills whereas the no-training group did not [t(18) = < 1]. EXPERIMENT 2 The first experiment demonstrated that young children can be taught to place emergency phone calls. In addition to knowing how to place emergency phone calls, it is extremely important for children to be able to identify the conditions under which such calls should be made. Previous work has shown that even though children report that they know when to perform their newly trained emergency phone skills, objective assessment does not support their reports (Jones, 1980). The present investigation trained children to discriminate the situations in which emergency dialing should be made. Method Subjects and Setting A discrimination task was administered to the 10 children in one of the behavioral training groups of Exp. 1. Three subjects performed the discrimination tasks well during baseline and were excluded from training, and another child no longer attended the school. The remaining children, three girls and three boys who ranged in age from 3 years to 4 years, 7 months (mean age = 4 years, 1 month) participated as subjects in Exp. 2. Training was conducted at the school in a room near the student's classroom. Assessment To assess the ability of children to discriminate when to make an emergency phone call, 30 pictures were presented daily to each child. 2 These 30 pictures, used for training and assessment purposes, depicted either z A v e r b a t i m a c c o u n t of the a s s e s s m e n t instructions can be obt a i ne d from the first a u t h o r

EMERGENCY DIALING SKILLS

515

14 BT

12

~10 -J

-Je if)

o

6

Z

_J