- Email: [email protected]
Therapist habituation and purported client progress in articulation therapy
JOURNAL OF COMMUNICATION
DISORDERS
9 (1976), 12%134
THERAPIST HABITUATION AND PURPORTED CLIENT PROGRESS IN ARTICULATION THERAPY STUART I. RITTERMAN The University of South Florida, Tampa, Florida 33620
ANTHONY
A. ZENNER
The University of South Florida, Tampa, Florida 33620
CAROL S. O’STEEN Hillsborough County Public Schools
Ten graduate student clinicians, in their final year of training, evaluated the tape recorded pretherapy performances of 10 /s/-defective children on the 68 /s/items of the McDonald Deep Test of Articularion (1964). Five of the 10 clinicians were randomly selected and assigned one of these /s/-defective children as part of their regular case load. The children were seen for 1 hr per week for a 5-week period. The clinicians were not informed as to the nature of the study or that the composition of their case load was in any way connected with the tapes that they had previously evaluated. The five remaining clinicians and children did not participate in therapy during this 5-week period. Immediately following this 5-week period, an additional tape recording was made of all 10 children’s responses to the/s/ items of the McDonald Deep Test (1964). Both pre- and posttreatment tapes were then evaluated by all 10 clinicians in independent listening sessions. Significant between-tape differences, indicative of both habituation and sensitization to client errors within the in-therapy clinicians only, are discussed in terms of their clinical applications.
Introduction Historically the field of speech pathology and audiology has placed its clinical focus on client pre- and postevalutation; effectiveness is usually validated in terms of pre- and posttest differences (Boone and Prescott, 1972). The usefulness of this type of difference score is at least in part a function of the accuracy and reliability of the examiner. Winitz (1969) has commented that “in evaluating the role of the examiner in articulation testing it is important to note that the examiner may introduce certain systematic biases that are not reflected in the reliability measures” (p. 241). Studies by Van Hattum (1960, 1970) report significantly high percentages of uncorrected misarticulations and unsuccessful carryover following dismissal from therapy. Van Hattum (1970) speculated that six factors have a possible bearing on this lack of carryover. They are (1) insufficient therapy, (2) superficial therapy, (3) the nature of the defect, (4) lack of flexibility in the therapy program, (5) poorly motivated children, and (6) no spread of effect in the therapy program (p. 306). Studies of reliability of judges by Henderson (1938), Sherman and Morrison (1955), Oyer (1959), Rintelman (1960), Winitz and Siegal (1961), Newman 0 American Elsevier Publishing
Company,
Inc.,
1976
129
130
STUART
I. RITTERMAN
et al
(1962)) Winitz (1963), and Irwin and Krafchick (1965) have reported high interand intraexaminer reliability. Winitz (1969) cautions, however, that “High judge reliability reported in many investigations should be accepted with reservations, as many studies report percentage indices” (p. 240). As Winitz points out, percentage indices may be spurious on a variety of counts. Although pre- and postevaluation of client performance would seem to be a fairly well developed aspect of the therapeutic procedure, the Van Hattum studies (1960, 1970) would seem to indicate that there is not necessarily a relationship between dismissal and cure. In this regard, Boone and Prescott (1972) asked: Do we know that what we do as clinicians does any good? While we are often convinced that our therapy has done some good-other listeners may lack the same conviction (p. 61).
Rephrasing this question we might ask, “do changes occur in the client’s mouth or in the clinician’s ear?” The findings reported in the studies by Van Hattum (1960, 1970) in particular seem strongly to suggest adaptation or habituation by the therapist to client errors. Coleman (1969) defined habituation in terms of generalization. He stated that: A common source When we know a over his head and other respects (p.
of error in our perceptions of others is the “halo effect” or “habituation”: person to be superior in some important respect, we may tend to put a halo through a process of generalization overrate his qualities or performance in 422).
Generalization is the tendency for a response, which has been conditioned to one stimulus, to become associated with other similar stimuli. Within the constraints of this definition, a relationship between performance and expected performance mediated by a halo effect, habituation, or generalization may be posited. Further consideration of the habituation problem necessitates discussion of the expectation aspects of the therapist successfully performing his role. As stated previously, the primary responsibility of therapists conducting articulation therapy is to alleviate and eliminate articulation errors. If he does not conform or produce within reasonable limits in ways that are perceived as consistent with role expectations , he is likely to encounter difficulties. Expected therapist performance thus produces an inner problem. This inner problem is discussed in terms of conceived and operative values by Coleman (1969). Coleman stated: Conceived values are conceptions of the ideal. For the most part, these are thevalues which the culture teaches and the ones most likely to be talked about in any discussion of “morality” or “ethics.” But conceived values, even though held with a good deal of intellectual conviction, sometimes have little practical influence on behavior. Operativevalues, on the other hand, are the criteria or value assumptions according to which action choices are actually made (p. 439).
THERAPIST
HABITUATION
AND PURPORTED
CLIENT
PROGRESS
131
For example, a therapist who has adequate training, experience, ethics, and the belief that client dismissal should be based upon error correction may be influenced by the need to show successful therapy; this too often has been determined solely by client dismissal. In selecting goals, in choosing means for reaching them, in resolving conflicts, the therapist is influenced at every turn by his conception of the preferable, the appropriate, the important, the good, and the desirable. Peterson and Beach (1967) have concluded: Other things being equal, an individual tends to select the course of action that seems to offer the greatest probability of success--as he defines success (p, 36).
The individual behaves in ways that are consistent with his concept of himself and tends to reject or distort incoming information that is inconsistent with or threatening to the self (Rogers, 1951). To add to the habituation problem, while assessing and determining goals, one inevitably uses past experiences and information, however limited or accurate, as a reference point. One of the most common mistakes is to assume that all members of a group will be like the examples one has known or has heard about. The prejudgment implies that the perceiver has formed judgments in advance about the characteristics and behavior of all members of a given group. Any member is then viewed as having these characteristics whether, in fact, he does or not. With our sources of information available to us we should be able to set goals and find good solutions. We are often thrown off the track, however, by our inability to see a problem for what it really is. Our information and experiences will fail to provide us with adequate goals and good solutions for therapy if we start out with false premises. Still another side of the habituation problem deals with cause-and-effect relationships or, more commonly, rationality. Rogers (1961) stated that, “the innermost core of man’s nature is basically rational and realistic” (p. 90). For example, the therapist who rationalizes may conclude that the purpose of successful therapy is to correct errors and the purpose of the therapist is to offer successful therapy; successful therapy then should result in error correction, and evaluation will reveal expected success. Methods In the present study 10 graduate student clinicians, in their final year of training, evaluated the tape recorded pretherapy performances of 10 /s/-defective children on the 68 Is/ items of the McDonald Deep Test of Articulation (1964) in independent listening sessions. Each item on this pretreatment tape was rated on a scale of from 0 (normal production) to 5 (grossly deviant production). Following this initial evaluation, 5 of the 10 clinicians were randomly selected
132
STUART
I. RITTERMAN
et al
and assigned one of the Is/ defective children as part of their regular case load. The children were seen for 1 hr per week for a 5-week period. Clinicians were not informed as to the nature of the study or that the composition of their case load was in any way connected with the tapes that they had evaluated previously. The remaining 5 clinicians and children did not participate in any therapy sessions during this 5-week period. Immediately following this 5week period, an additional tape recording was made of all 10 children’s responses to the 68 Is/ items of the McDonald Deep Test (1964). Both the pre- and posttreatment tapes were then evaluated by all 10 clinicians in independent listening sessions. The presentation order of the tapes was randomized; the clinicians were given no additional information regarding the nature of the study. Results In order to ensure that there were no pretreatment between-group differences between those clinicians who were to be involved in therapy and those not to participate in therapy, an analysis of variance was performed on all clinicians’ first evaluation of the pretreatment tape. Results of this analysis yielded no significant (p > 0.01) between-group differences. The results of the posttreatment comparisons will be discussed with respect to both absolute and relative deviations from the base-line judgments obtained from the first evaluation of the pretreatment tape. A relative measure takes into account the directionality of the deflections from baseline. That is, a relative measure will indicate if clinicians deviate in a given direction from their original observations (positive deflections would be indicative of increased sensitivity to client errors; negative deflections from baseline would be indicative of habituation to client errors). Absolute comparisons, on the other hand, give only an indication of the gross variability with no regard to the direction of the deviations from baseline. There were no significant (p > 0.05) relative differences between clinicians participating in therapy and those not participating in therapy. An analysis of variance of the relative differences between clinician groups, of the first and second evaluations of the pretreatment tape, as well as the differences between their evaluations of the pre- and posttreatment tapes, yielded F ratios less than unity. These results would seem to indicate that neither habituation nor increased sensitivity to client errors occurred as a group phenomenon in either the in-therapy or nontherapy clinician group. When examining the absolute deflections from baseline, however, significant differences (p < 0.01) were observed between clinicians participating in therapy and those not participating in therapy. An analysis of variance (ANOVA) of the absolute differences between clinician groups on their first and second evaluations of the pretreatment tape, as well as the differences between their evaluations of the pre- and posttreatment tapes, yielded significant F ratios indicative of significantly
THERAPIST
HABITUATION
AND
PURPORTED
CLIENT
PROGRESS
133
greater variability on the part of the in-therapy clinicians. A between-clinician group F ratio of 8.05 @I < 0.01) was observed in the comparison of the first and second evaluation of the pretreatment tape. A between-clinician group F ratio of 4.29 @ < 0.01) was observed in the comparison of the pre- and posttreatment tapes. In summary, there were no significant relative differences between the two clinician groups irrespective of the tapes evaluated. Clinicians engaged in therapy, however, demonstrated significant absolute differences both between their first and second evaluations of the pretreatment tape and between their evaluation of the pre- and posttreatment tape. No significant between-tape differences were observed, however, in the evaluation of the nontherapy clinicians. In addition, no differences were observed within the in-therapy clinicians in their evaluations of the children they treated and those they did not. Discussion The failure to observe significant relative differences between the two clinician groups would seem to indicate that neither habituation nor increased sensitivity to client errors occurred as a group phenomenon. To some extent, this may be attributable to a cancelling out process. That is, some of the clinician’s posttreatment evaluations of the tapes revealed perceived improvement (habituation) in comparison with their pretreatment evaluations. Others, however, demonstrated an increase in perceived errors. For the most part, however, there was no directionality in the variance of either group from their base-line pretreatment evaluations . When examining the absolute differences between clinician groups, however, the in-therapy clinicians perceived significantly greater changes than did the nontherapy clinicians. The nontherapy clinicians demonstrated very little variability, either between or within clinicians, in their evaluations of all pre- and posttreatment tapes. The in-therapy clinicians, however, showed little or no correspondence (either within or between clinicians) in their two evaluations of the pretreatment tape. These results would seem to suggest that the student clinician involved in a therapy practicum is not yet sufficiently stable to perform independent evaluations of the performance status of the client. There were no significant absolute differences between the two clinician groups’ posttreatment evaluations of the pre- and posttreatment tapes. When the evaluations of the base-line tapes were compared with those of the tape prepared after the 5-week treatment period, however, significant differences were observed only within the in-therapy clinicians. Again, these differences were nondirectional. That is, some of the clients were evaluated more severely than they were on the base-line tape; others, however, were evaluated as improved. There was, in addition, poor interclinician agreement as to the direction of the change for any given client.
134
STUART
I. RITTERMAN
et al.
To some extent, these findings may be explained by considering the results of temporal reliability studies. No significant differences were observed in either of the two clinician groups’ evaluations of the two posttreatment tapes. These evaluations were made on the same day. All significant differences were related to differences between the base-line evaluations and the posttreatment evaluations. These evaluations were separated by a 5-week interval. In this respect Winitz ( 1969) stated, “Variations in articulation scores are minimal over short time intervals, but increase when longer time intervals between test-retest situations are introduced” (p. 239). While neither adaptation nor increased sensitivity to client errors appeared to exist as a group phenomenon, increased variability in clinicians’ judgments of articulatory responses as a function of participation in therapy was observed. This finding would seem to suggest that student clinicians involved in therapy are not sufficiently stable in their evaluations of their clients’ performances to render an adequate independent diagnosis. It would seem to follow, therefore, that during the evaluative process, student clinicians should be closely supervised. It should be noted, however, that to date, it has yet to be demonstrated that experienced clinicians are any more stable in their evaluations of pre- and posttherapy articulation performance. The results of the studies by Van Hattum (1960, 1970) would seem to indicate that this area bears further investigation. References Boone, D. R., Prescott, T. E. Content and sequence analysis of speech and hearing therapy. ASHA, 1972, 14, 58-62. Coleman, I. C., Psychology and ESfective Behavior. Glenview, Illinois: Scott, Foresman & Co., 1969. Henderson, F. M. Accuracy in testing the articulation of speech sounds. J. E&c. Res., 1938, 31, 348-356. Irwin, R. B., Krafchick, I. P. An audio-visual test for evaluating the ability to recognize phonetic errors. J. Speech Hearing Res., 1965, 8, 281-290. McDonald, E. T. A deep tesr ofarficulation. Pittisburgh. Stanwix House, Inc., 1964. Newman, P. W. Speech impaired? ASHA, 1961, 3, 9-10. Oyer, H. J., Speech error recognition ability. J. Speech Hearing Dis. 1959, 24, 391-394. Peterson, C. R., Beach, L. R. Man as an intuitive statistician. Psychol. Bull., 1967, 68, 29-46. Rintelman, W. F. Changes in the articulatory responses of preschool children traced through four successive six-week periods. Ph.D. dissertation, Indiana University, 1960. Rogers, C. R. Client cenrered therapy. Boston: Houghton Mifflin, 1951. Sherman, D., Morrison, S. Reliability of individual ratings of the severity of defective articulation. J. Speech Hearing Dis., 1955, 20, 352-358. Van Hattum, R. J. Unpublished paper. In: Van Hattum, R. J., Clinical speech in the schools. Springfield, Ill.: Charles C Thomas, 1960. Van Hattum, R. J. Clinical speech in the schools. Springfield, Ill.: Charles C Thomas, 1970. Winitz, H. Temporal reliability in articulation testing. J. Speech Hearing Dis., 1963,28,247-25 1. Winitz, H. Articulafory acquisition and behavior. New York: Appleton-Century-Crofts, 1969. Winitz, H., Siegel, Cl. M. Interest variability in articulation testing. Unpublished manuscript, 1961.
DISORDERS
9 (1976), 12%134
THERAPIST HABITUATION AND PURPORTED CLIENT PROGRESS IN ARTICULATION THERAPY STUART I. RITTERMAN The University of South Florida, Tampa, Florida 33620
ANTHONY
A. ZENNER
The University of South Florida, Tampa, Florida 33620
CAROL S. O’STEEN Hillsborough County Public Schools
Ten graduate student clinicians, in their final year of training, evaluated the tape recorded pretherapy performances of 10 /s/-defective children on the 68 /s/items of the McDonald Deep Test of Articularion (1964). Five of the 10 clinicians were randomly selected and assigned one of these /s/-defective children as part of their regular case load. The children were seen for 1 hr per week for a 5-week period. The clinicians were not informed as to the nature of the study or that the composition of their case load was in any way connected with the tapes that they had previously evaluated. The five remaining clinicians and children did not participate in therapy during this 5-week period. Immediately following this 5-week period, an additional tape recording was made of all 10 children’s responses to the/s/ items of the McDonald Deep Test (1964). Both pre- and posttreatment tapes were then evaluated by all 10 clinicians in independent listening sessions. Significant between-tape differences, indicative of both habituation and sensitization to client errors within the in-therapy clinicians only, are discussed in terms of their clinical applications.
Introduction Historically the field of speech pathology and audiology has placed its clinical focus on client pre- and postevalutation; effectiveness is usually validated in terms of pre- and posttest differences (Boone and Prescott, 1972). The usefulness of this type of difference score is at least in part a function of the accuracy and reliability of the examiner. Winitz (1969) has commented that “in evaluating the role of the examiner in articulation testing it is important to note that the examiner may introduce certain systematic biases that are not reflected in the reliability measures” (p. 241). Studies by Van Hattum (1960, 1970) report significantly high percentages of uncorrected misarticulations and unsuccessful carryover following dismissal from therapy. Van Hattum (1970) speculated that six factors have a possible bearing on this lack of carryover. They are (1) insufficient therapy, (2) superficial therapy, (3) the nature of the defect, (4) lack of flexibility in the therapy program, (5) poorly motivated children, and (6) no spread of effect in the therapy program (p. 306). Studies of reliability of judges by Henderson (1938), Sherman and Morrison (1955), Oyer (1959), Rintelman (1960), Winitz and Siegal (1961), Newman 0 American Elsevier Publishing
Company,
Inc.,
1976
129
130
STUART
I. RITTERMAN
et al
(1962)) Winitz (1963), and Irwin and Krafchick (1965) have reported high interand intraexaminer reliability. Winitz (1969) cautions, however, that “High judge reliability reported in many investigations should be accepted with reservations, as many studies report percentage indices” (p. 240). As Winitz points out, percentage indices may be spurious on a variety of counts. Although pre- and postevaluation of client performance would seem to be a fairly well developed aspect of the therapeutic procedure, the Van Hattum studies (1960, 1970) would seem to indicate that there is not necessarily a relationship between dismissal and cure. In this regard, Boone and Prescott (1972) asked: Do we know that what we do as clinicians does any good? While we are often convinced that our therapy has done some good-other listeners may lack the same conviction (p. 61).
Rephrasing this question we might ask, “do changes occur in the client’s mouth or in the clinician’s ear?” The findings reported in the studies by Van Hattum (1960, 1970) in particular seem strongly to suggest adaptation or habituation by the therapist to client errors. Coleman (1969) defined habituation in terms of generalization. He stated that: A common source When we know a over his head and other respects (p.
of error in our perceptions of others is the “halo effect” or “habituation”: person to be superior in some important respect, we may tend to put a halo through a process of generalization overrate his qualities or performance in 422).
Generalization is the tendency for a response, which has been conditioned to one stimulus, to become associated with other similar stimuli. Within the constraints of this definition, a relationship between performance and expected performance mediated by a halo effect, habituation, or generalization may be posited. Further consideration of the habituation problem necessitates discussion of the expectation aspects of the therapist successfully performing his role. As stated previously, the primary responsibility of therapists conducting articulation therapy is to alleviate and eliminate articulation errors. If he does not conform or produce within reasonable limits in ways that are perceived as consistent with role expectations , he is likely to encounter difficulties. Expected therapist performance thus produces an inner problem. This inner problem is discussed in terms of conceived and operative values by Coleman (1969). Coleman stated: Conceived values are conceptions of the ideal. For the most part, these are thevalues which the culture teaches and the ones most likely to be talked about in any discussion of “morality” or “ethics.” But conceived values, even though held with a good deal of intellectual conviction, sometimes have little practical influence on behavior. Operativevalues, on the other hand, are the criteria or value assumptions according to which action choices are actually made (p. 439).
THERAPIST
HABITUATION
AND PURPORTED
CLIENT
PROGRESS
131
For example, a therapist who has adequate training, experience, ethics, and the belief that client dismissal should be based upon error correction may be influenced by the need to show successful therapy; this too often has been determined solely by client dismissal. In selecting goals, in choosing means for reaching them, in resolving conflicts, the therapist is influenced at every turn by his conception of the preferable, the appropriate, the important, the good, and the desirable. Peterson and Beach (1967) have concluded: Other things being equal, an individual tends to select the course of action that seems to offer the greatest probability of success--as he defines success (p, 36).
The individual behaves in ways that are consistent with his concept of himself and tends to reject or distort incoming information that is inconsistent with or threatening to the self (Rogers, 1951). To add to the habituation problem, while assessing and determining goals, one inevitably uses past experiences and information, however limited or accurate, as a reference point. One of the most common mistakes is to assume that all members of a group will be like the examples one has known or has heard about. The prejudgment implies that the perceiver has formed judgments in advance about the characteristics and behavior of all members of a given group. Any member is then viewed as having these characteristics whether, in fact, he does or not. With our sources of information available to us we should be able to set goals and find good solutions. We are often thrown off the track, however, by our inability to see a problem for what it really is. Our information and experiences will fail to provide us with adequate goals and good solutions for therapy if we start out with false premises. Still another side of the habituation problem deals with cause-and-effect relationships or, more commonly, rationality. Rogers (1961) stated that, “the innermost core of man’s nature is basically rational and realistic” (p. 90). For example, the therapist who rationalizes may conclude that the purpose of successful therapy is to correct errors and the purpose of the therapist is to offer successful therapy; successful therapy then should result in error correction, and evaluation will reveal expected success. Methods In the present study 10 graduate student clinicians, in their final year of training, evaluated the tape recorded pretherapy performances of 10 /s/-defective children on the 68 Is/ items of the McDonald Deep Test of Articulation (1964) in independent listening sessions. Each item on this pretreatment tape was rated on a scale of from 0 (normal production) to 5 (grossly deviant production). Following this initial evaluation, 5 of the 10 clinicians were randomly selected
132
STUART
I. RITTERMAN
et al
and assigned one of the Is/ defective children as part of their regular case load. The children were seen for 1 hr per week for a 5-week period. Clinicians were not informed as to the nature of the study or that the composition of their case load was in any way connected with the tapes that they had evaluated previously. The remaining 5 clinicians and children did not participate in any therapy sessions during this 5-week period. Immediately following this 5week period, an additional tape recording was made of all 10 children’s responses to the 68 Is/ items of the McDonald Deep Test (1964). Both the pre- and posttreatment tapes were then evaluated by all 10 clinicians in independent listening sessions. The presentation order of the tapes was randomized; the clinicians were given no additional information regarding the nature of the study. Results In order to ensure that there were no pretreatment between-group differences between those clinicians who were to be involved in therapy and those not to participate in therapy, an analysis of variance was performed on all clinicians’ first evaluation of the pretreatment tape. Results of this analysis yielded no significant (p > 0.01) between-group differences. The results of the posttreatment comparisons will be discussed with respect to both absolute and relative deviations from the base-line judgments obtained from the first evaluation of the pretreatment tape. A relative measure takes into account the directionality of the deflections from baseline. That is, a relative measure will indicate if clinicians deviate in a given direction from their original observations (positive deflections would be indicative of increased sensitivity to client errors; negative deflections from baseline would be indicative of habituation to client errors). Absolute comparisons, on the other hand, give only an indication of the gross variability with no regard to the direction of the deviations from baseline. There were no significant (p > 0.05) relative differences between clinicians participating in therapy and those not participating in therapy. An analysis of variance of the relative differences between clinician groups, of the first and second evaluations of the pretreatment tape, as well as the differences between their evaluations of the pre- and posttreatment tapes, yielded F ratios less than unity. These results would seem to indicate that neither habituation nor increased sensitivity to client errors occurred as a group phenomenon in either the in-therapy or nontherapy clinician group. When examining the absolute deflections from baseline, however, significant differences (p < 0.01) were observed between clinicians participating in therapy and those not participating in therapy. An analysis of variance (ANOVA) of the absolute differences between clinician groups on their first and second evaluations of the pretreatment tape, as well as the differences between their evaluations of the pre- and posttreatment tapes, yielded significant F ratios indicative of significantly
THERAPIST
HABITUATION
AND
PURPORTED
CLIENT
PROGRESS
133
greater variability on the part of the in-therapy clinicians. A between-clinician group F ratio of 8.05 @I < 0.01) was observed in the comparison of the first and second evaluation of the pretreatment tape. A between-clinician group F ratio of 4.29 @ < 0.01) was observed in the comparison of the pre- and posttreatment tapes. In summary, there were no significant relative differences between the two clinician groups irrespective of the tapes evaluated. Clinicians engaged in therapy, however, demonstrated significant absolute differences both between their first and second evaluations of the pretreatment tape and between their evaluation of the pre- and posttreatment tape. No significant between-tape differences were observed, however, in the evaluation of the nontherapy clinicians. In addition, no differences were observed within the in-therapy clinicians in their evaluations of the children they treated and those they did not. Discussion The failure to observe significant relative differences between the two clinician groups would seem to indicate that neither habituation nor increased sensitivity to client errors occurred as a group phenomenon. To some extent, this may be attributable to a cancelling out process. That is, some of the clinician’s posttreatment evaluations of the tapes revealed perceived improvement (habituation) in comparison with their pretreatment evaluations. Others, however, demonstrated an increase in perceived errors. For the most part, however, there was no directionality in the variance of either group from their base-line pretreatment evaluations . When examining the absolute differences between clinician groups, however, the in-therapy clinicians perceived significantly greater changes than did the nontherapy clinicians. The nontherapy clinicians demonstrated very little variability, either between or within clinicians, in their evaluations of all pre- and posttreatment tapes. The in-therapy clinicians, however, showed little or no correspondence (either within or between clinicians) in their two evaluations of the pretreatment tape. These results would seem to suggest that the student clinician involved in a therapy practicum is not yet sufficiently stable to perform independent evaluations of the performance status of the client. There were no significant absolute differences between the two clinician groups’ posttreatment evaluations of the pre- and posttreatment tapes. When the evaluations of the base-line tapes were compared with those of the tape prepared after the 5-week treatment period, however, significant differences were observed only within the in-therapy clinicians. Again, these differences were nondirectional. That is, some of the clients were evaluated more severely than they were on the base-line tape; others, however, were evaluated as improved. There was, in addition, poor interclinician agreement as to the direction of the change for any given client.
134
STUART
I. RITTERMAN
et al.
To some extent, these findings may be explained by considering the results of temporal reliability studies. No significant differences were observed in either of the two clinician groups’ evaluations of the two posttreatment tapes. These evaluations were made on the same day. All significant differences were related to differences between the base-line evaluations and the posttreatment evaluations. These evaluations were separated by a 5-week interval. In this respect Winitz ( 1969) stated, “Variations in articulation scores are minimal over short time intervals, but increase when longer time intervals between test-retest situations are introduced” (p. 239). While neither adaptation nor increased sensitivity to client errors appeared to exist as a group phenomenon, increased variability in clinicians’ judgments of articulatory responses as a function of participation in therapy was observed. This finding would seem to suggest that student clinicians involved in therapy are not sufficiently stable in their evaluations of their clients’ performances to render an adequate independent diagnosis. It would seem to follow, therefore, that during the evaluative process, student clinicians should be closely supervised. It should be noted, however, that to date, it has yet to be demonstrated that experienced clinicians are any more stable in their evaluations of pre- and posttherapy articulation performance. The results of the studies by Van Hattum (1960, 1970) would seem to indicate that this area bears further investigation. References Boone, D. R., Prescott, T. E. Content and sequence analysis of speech and hearing therapy. ASHA, 1972, 14, 58-62. Coleman, I. C., Psychology and ESfective Behavior. Glenview, Illinois: Scott, Foresman & Co., 1969. Henderson, F. M. Accuracy in testing the articulation of speech sounds. J. E&c. Res., 1938, 31, 348-356. Irwin, R. B., Krafchick, I. P. An audio-visual test for evaluating the ability to recognize phonetic errors. J. Speech Hearing Res., 1965, 8, 281-290. McDonald, E. T. A deep tesr ofarficulation. Pittisburgh. Stanwix House, Inc., 1964. Newman, P. W. Speech impaired? ASHA, 1961, 3, 9-10. Oyer, H. J., Speech error recognition ability. J. Speech Hearing Dis. 1959, 24, 391-394. Peterson, C. R., Beach, L. R. Man as an intuitive statistician. Psychol. Bull., 1967, 68, 29-46. Rintelman, W. F. Changes in the articulatory responses of preschool children traced through four successive six-week periods. Ph.D. dissertation, Indiana University, 1960. Rogers, C. R. Client cenrered therapy. Boston: Houghton Mifflin, 1951. Sherman, D., Morrison, S. Reliability of individual ratings of the severity of defective articulation. J. Speech Hearing Dis., 1955, 20, 352-358. Van Hattum, R. J. Unpublished paper. In: Van Hattum, R. J., Clinical speech in the schools. Springfield, Ill.: Charles C Thomas, 1960. Van Hattum, R. J. Clinical speech in the schools. Springfield, Ill.: Charles C Thomas, 1970. Winitz, H. Temporal reliability in articulation testing. J. Speech Hearing Dis., 1963,28,247-25 1. Winitz, H. Articulafory acquisition and behavior. New York: Appleton-Century-Crofts, 1969. Winitz, H., Siegel, Cl. M. Interest variability in articulation testing. Unpublished manuscript, 1961.