Charles H. Hill

SUMMER remedial reading programs are very popular and expensive. The expenditure of the tax dollar would seem to beg for some empirical justification. Although most educators who conduct summer reading programs claim astounding success, and most parents, perhaps wishfully, give glowing testimonials of the great improvement in their offspring, there is a paucity of research evidence demonstrating the results of short-term summer programs.

Moderate gains in reading level following a summer reading program for first, second, and third graders were reported by Arron, Callaway, Hicks, and Simpson (1969) in an end of summer evaluation. Reporting longitudinal effects of a summer televised reading program, Humphry (1968) found that the summer assistance helped compensate for the summer losses of elementary children which many teachers recognize as inevitable. Early (1969) reviewed the literature measuring success in reading programs. She noted significant gains for massive, long-term programs. Most of the studies she reported, however, failed to include either a control group or a longitudinal follow-up.

Studies of the longitudinal benefits of remedial instruction have been done by Barlow (1965), Bliesmer (1962), Buerger (1968) and Lovell, Johnson and Platts (1962). These investigations share a common finding that progress accelerates rapidly during remedial instruction and returns to the previous sub-normal pace when remedial treatment ceases.

The results of these studies imply that a district or community will not significantly improve the reading of children who have problems by instituting summer programs, compensatory programs which are funded for a few years and then abandoned, or programs, relatively unconnected with regular classroom activities, giving an hour or two a week of remediation to a select few in a special room (frequently the boiler room or back-stage). The ever present abundance of testimonials notwithstanding, real, measurable, and enduring progress for "remedial" readers seems to require a massive, long-term commitment which is related closely to the on-going classroom curriculum.

A summer reading program which may be typical Of the moderately funded, short-duration, detached from the regular school program type of remedial activity was recently conducted by the writer. The Bureau of Indian Affairs funded a six-week remedial reading program for the Nez Perce Indian Reservation. Following is a report of this activity, including a follow-up evaluation and possible implications for future similar activities.

All pupils were American Indian, living on a rural Indian reservation. The total enrollment, those attending two or more sessions, included 15 boys and four girls, spanning grades four through eight. Thirty-six students were referred by teachers and school administration. Criteria for referral were: Indian descent, estimated reading one or more grade levels below grade placement, not certified mentally retarded by a qualified psychologist.

Each referred student was given a diagnostic test battery which included: WISC Performance Scale, Peabody Picture Vocabulary Test, Wepman Test of Auditory Discrimination, Durrell Analysis of Reading Difficulty, and the California Reading Test.

The referred students tested displayed relatively average scores on the WISC Performance Scale. Scores on the PPVT lagged 8 to 12 points below WISC scores. More than half of the students scored below age norms on the Wepman Test. Individual reading tests revealed a population which performed at or above grade level in oral pronunciation. This may be attributed to the Phonetic Keys to Reading program used in the primary grades. Silent reading comprehension lagged about two years below grade level and auditory comprehension of material read to them lagged three to four years below grade level. Vocabulary level seemed to be correlated with the silent reading comprehension level. Basic weaknesses included knowledge or recognition of affixes, use of context skills, knowledge of study skills, and inability of cope with interpretation questions or literary devices.

Remedial sessions were held three hours daily, Monday through Fri day, for six weeks. The average daily attendance was 13.6. Nine students attended 75% or more sessions. Attendance was high for the first four weeks but dropped off sharply the final two weeks of the program.

Classes were held in the community center. This proved to be a liability to the program because of the proximity of friends and entertainment facilities. A VISTA worker picked up the children at home and returned them after class. The VISTA worker also took the regular attenders to a public swimming pool, twice a week, at program expense.

Two American Indians were hired as teachers. One was a recent elementary education graduate and the other a senior in secondary education. Training in the use of materials was given to the teachers, as well as consultation in program planning and test interpretation. The program director made a tentative individualized program for each child, using test data. Although the intention was to allow some self-selection and self-pacing within the confines of each individual program, students were so lacking in experience with self-sustained effort, self-evaluation, and decision-making that chaos had to be avoided by the rigid scheduling of daily activities for each child.

Materials included SRA, Sullivan, Barnell-Loft, Bobbs-Merrill, and EDL programmed materials. This list is not exhaustive. These materials centered on reading, but also provided arithmetic and English activities. A 100 volume paper-back library was purchased and heavily used. In retrospect, the success of a given set of materials seemed to depend upon group acceptance or rejection. This, together with the distractability of the students, leads to a conclusion that study carrells or some other form of isolation would have enhanced the program.

The students were self-selected into control and experimental groups, using the students referred as the population. Those who attended two or more sessions were treated as the experimental group, those declining attendance as the control. The original experimental group was 19, control was 17. Only fourth, fifth, and sixth graders were retained in either group. Those with 60% or better attendance were used for the experimental group. Five members of the control group and one member of the experimental group moved from the area prior to the post-test. An analysis of the final groups is presented in Table 1.

All students were administered the California Reading Test, Intermediate Battery, Form Y, as a pre-test during the third week of May. The same test, Form X, was administered as a post-test during the last week of September, after school had been in session one month. Test results are reported in Table 2.

A one-tailed test of significance using the Mann-Whitney U Test was applied to the null hypothesis that there would be no significant difference between the mean grade equivalent score changes in vocabulary comprehension, and total reading of the experimental and control groups. For the vocabulary, the hypothesis was rejected at the .025 level of confidence. The hypothesis was also rejected for the total reading score at the .05 level of confidence. Although the control group scored a measurably larger regressive score in comprehension, the difference was not significant and the hypothesis was accepted.

Table 1

Experimental Control

Table 2

Mean Change in Grade Quivalent Scores

* significant at .05

**significant at .025

This research design suffers a number of maladies common to research in education, especially when small numbers and mobile groups are involved. Yet it is with similar groups that we spend massive amounts of money. Some form of empirical evaluation must be attempted, however weak the design may be.

The inability to randomize was expected to introduce student attitude as an unmeasured variable. While student attitude no doubt influenced the data, it might be argued that there was very little difference in attitude toward reading between the two groups. Most of the experimental students attended at the insistence of parents or some other influential adult. In fact, very few would have attended of their own volition. Attitudes toward the program worsened during the final two weeks when the experimental group was "forced" in some way to attend.

The regressive comprehension score of the experimental group may be explained by a combination of negative attitude and testing conditions. Two girls, whose attitude toward the tester-director was very negative, scored more than three grade levels below the pre-test. With these scores removed, there was measurable, but not significant, comprehension score difference in favor of the experimental group. While the control group was given a physical activity break mid-way through the post-test, the experimental group, tested the following day, could not have the break. The length of the session, combined with the negative approach to the testing situation may have caused the extreme regressive score. At best, making inferences from one testing instrument given to "disadvantaged" children is risky.

Nevertheless, the findings of significance between the vocabulary and total reading scores indicates that a summer program may be worth continued consideration and evaluation. Apparently an important benefit is the prevention of regression in reading skills during the summer months. Even if difference in attitude were an influential variable, the opportunity for guided practice must be provided.

Further longitudinal study might be interesting but would be an exercise in futility. The N of both groups continues to decline. If the line of reasoning established in the review of literature is valid, loss of significant differences is expected unless remediation is continued. What happens beyond this point one month after school starts is the responsibility of the regular school program, not a previous program which filled a specific short-term need.

Arron, Callaway, Hicks, Simpson. "Reading Achievement in a Summer Reading Program." Elementary English, 1969, 42, 56-59.

Barlow, Bruce. "The Long-Term Effect of Remedial Reading Instruction." The Reading Teacher, 1965, 18, 581-586.

Bliesmer, Emery P. "Evaluating Progress in Remedial Reading Programs." The, Reading Teacher, 1962, 15, 344-350.

Buerger, Theodore A. "A Follow-Up of Remedial Reading Instruction." The Reading Teacher, 1968, 21, 329-334.

Early, Margaret J. "What Does Research in Reading Reveal About Successful Reading Programs." English Journal, 1969, 38, 534-546.

Humphry, Jack W. "The Effect of a Summer Television Reading Program on the Reading Achievement of Children." Forging Ahead In Reading, International Reading Association, Newark, Delaware, 1968.

Lovell, K., Johnson, E., and Platts, D. "A Summary of a Study of the Reading Ages of Children Who Had Been Given Remedial Teaching." The British Journal of Educational Psychology, 1962, 32, 66-71.