Educational Psychology, Vol. 23, No. 5, December 2003

Primary Teachers’ Attitudes Toward the Direct Instruction Construct

MEAGAN S. DEMANT & GREGORY C. R. YATES, University of South Australia, Australia

In this survey we sought to investigate the extent to which primary school teachers working in Adelaide’s northern suburbs (mainly lower SES) would relate to direct instruction as a viable teaching method in their professional work. Through approaches in school staffrooms, 150 questionnaires were distributed and 58 of these were returned via mail. A Likert-scale was used with five positive and six negative items, and a single factor resolution was evident. It was possible to identify 11 (19%) respondents exhibiting varying degrees of negative attitude, and 47 (81%) exhibiting varying degrees of positive attitude. Attitudes to direct instruction correlated positively with teachers’ years of experience (r ⫽ 0.34), and with a checklist measure tapping actual knowledge of the components of direct instruction as described by Rosenshine (r ⫽ 0.63). Female teachers reported more positive attitudes than male teachers. Item analysis indicated a consistent pattern of generally positive orientation towards direct instruction, except in the case of one item, “Direct instruction is an effective method with all students,” which elicited an agreement level of only 39%.

ABSTRACT

In this project we sought to investigate how teachers at the primary (that is, elementary) school level would relate to the construct of “direct instruction.” As a term describing a defined pattern or style of teaching, this construct appears to evoke remarkably different associations in different people. Within this project we sought to tap into these different perceptions, and investigate teacher traits that might be associated with such variations. Rosenshine and Meister (1995) have noted that research-oriented writers within the field of education have employed the construct of direct instruction in five distinct ways: 1. as a generic concept for all forms of teacher-initiated classroom activity 2. to refer to instructional procedures correlated with learning gains in processproduct classroom research (that is, teacher effects research) ISSN 0144-3410 print; ISSN 1469-046X online/03/050483-07  2003 Taylor & Francis Ltd DOI: 10.1080/0144341032000123741

484

M. S. Demant & G. C. R. Yates

3. as instructional procedures used to teach cognitive information processing strategies 4. as a curriculum product, most keenly represented by the DISTAR approach 5. as a pejorative term referring to instances where teacher-centred activity is assumed to conflict with student-centred needs or other active developmental opportunities Rosenshine and Meister (1995) suggested a considerable overlap in the way these meanings are articulated. However, it is apparent that at least two of these (notably items 2 and 5) are in clear antagonism. We wondered if the tension between these two positions represents an attitudinal dimension associated with the manner in which individual teachers reflect upon their professional role. Such a dimension is clearly mirrored in the educational research literature. Researchers have often sought to describe teaching methods, philosophies, and goals in terms of tension between directive and less systematically teacher-led approaches (Baumann, 1988; Chall, 2000; Cobb, 1988; Delpit, 1988; Speigel, 1992). However, the extent to which practicing teachers will also find the term direct instruction a meaningful anchor point for reflecting upon professional practice is unclear. We decided to conduct a survey into primary teachers’ attitudes towards the direct instruction construct. We were interested in how many teachers would express their attitudes in positive or negative directions, and this was measured by asking for levels of agreement with a series of statements. Following on from the attitudinal statements, teachers were also asked to tick a checklist of 20 items, to tap into their knowledge of the concept of direct instruction as it tends to be used within modern teacher education courses: for instance, the components as described by significant reviewers such as Rosenshine and Stevens (1986). Method Participants Teachers were approached with a personal invitation, which took place mainly in school staffrooms, in both Catholic and state schools. Each teacher was asked to take a questionnaire (which took around 15 minutes to complete) and return it to the researchers in a supplied envelope, either at a later point in time, or via the school courier service. Names were not recorded. Around 150 questionnaires were distributed and 58 were returned. The seven schools were all in the northern suburbs of Adelaide, South Australia, in predominantly lower SES districts. Of the respondents it was found that 10 identified themselves as junior primary, 34 as upper primary; 20 were male and 38 female; and the median length of teaching experience was 15.0 years. Instrument The actual questionnaire used is available as a rich text file accessible from the website http://www.unisanet.unisa.edu.au/edpsych/quests. The following were measured: 1. Attitudinal responses: We devised five statements that reflected positive views, and six statements that reflected negative views. Teachers were asked to use a seven-point Likert scale ranging from very strongly disagree (1), to very strongly agree (7), with no real opinion being expressed as the midpoint (4).

Teacher Attitudes to Direct Instruction

485

2. Knowledge assessment: A 20-item checklist was used, where teachers were asked to tick those elements they considered key components of direct instruction. Of the 20 items, 17 were deemed to represent components of direct instruction as described by Rosenshine and Stevens (1986). The remaining three items were inserted as foils (for example, “Teacher allows students to discover new information for themselves”). 3. Knowledge of researchers: Teachers were asked if they could name the occupations of 20 real people. Some of these were famous individuals (such as Cathy Freeman and Ben Chifley), but seven were people associated with educational research in the area of direct instruction (Barak Rosenshine, Tom Good, David Berliner, Ellen Gagne, Ann Brown, Jere Brophy, and Michael Pressley). Thus, a score out of seven was used to index the extent to which each teacher was aware of direction instruction researcher workers. 4. Teacher characteristics: Gender, number of years of teaching experience, and level of teaching in terms of the junior or upper primary division, were requested.

Results A principal components procedure on the 11 attitude statement items revealed that all items loaded strongly onto a single factor. Thus, it was meaningful to combine these items to form a single attitude score with a possible range from 11 to 77. The actual range was 25 to 75, with a median of 57, mean of 53.5, and a standard deviation of 13.8. Skewness and kurtosis coefficients indicated acceptable curve properties, and the internal reliability (alpha coefficient) was 0.92. It can be noted that the natural midpoint on this scale (that is, consistent responding on the neutral point) would be 44. Inspection of the item means revealed that teachers responded in the positive direction on 10 of the 11 items. The one item where the mean dropped below four was on the statement “Direct instruction is an effective method with all students” (see Table I). If we regard 44 as the natural midpoint then it was apparent that 47 teachers (81%) exhibited scores above this level and can be said to possess positive attitudes, and 11 teachers (19%) were below this point and can be said to display negative attitudes. No respondent actually scored on the midline point. Attitude scores were found to correlate positively with years of teaching experience (r ⫽ 0.34, P ⬍ 0.01). Although the junior/upper primary teacher variable did not relate to attitude scores, it was found that the female teachers exhibited more positive attitudes than the male teachers [means of 57.0 and 47.9, F (1, 56) ⫽ 7.2, P ⫽ 0.01, effect size Cohen’s d of 0.74]. Through additional regression and ANOVA analyses we established that gender and experience effects did not interact significantly. The relationships are expressed in Fig. 1 by dividing the sample into two groups at the 15 years’ experience point. In Fig. 1, the boxplot procedure depicts medians and quartile ranges, and a reference line is drawn at the natural midpoint on the attitude scale. On the knowledge checklist, the mean number of direct instruction elements ticked was 10.1 out of 17. It was found that the knowledge score correlated strongly with the attitude score (r ⫽ 0.63, P ⫽ 0.001). Chi-square procedures (P ⬍ 0.01 in each case) were then used to identify specific knowledge items which clearly discriminated teachers with positive and negative attitudes. These five items were: 1. teacher states lesson objectives 2. teacher asks many questions

1.8

4.1

12.9

1.7

4.8

53.8

1.8 1.5 1.8 1.7 1.7 1.4 1.1 1.6 1.8

SD

5.4 5.4 4.6 4.5 3.7 5.2 5.5 5.4 5.1

Mean

81

46

66

81 76 66 60 39 76 86 81 76

Pos %

Notes: Item means are all expressed in the positive direction, where 7 represents strong positive attitude, 4 is the neutral point, and 1 represents strong negative attitude. (R) refers to items reversed for scoring purposes. On overall tally, a score of 44 represents the natural midpoint. The column headed “Pos %” refers to the percentage of the sample who responded in the positive direction on each specific item. In other words, on the second item, 76% of the sample disagreed with the negative attitude expressed by the statement on that item

Total Attitude Score

Direct instruction is a useful teaching method for teaching basic skills Direct instruction, as a teaching method, is harmful to children’s mental development (R) Receiving teaching based on direct instruction deprives children of the ability to think for themselves (R) Direct instruction can be used to teach thinking skills Direct instruction is a highly effective teaching method with all students Direct instruction is useless in helping students to really understand things (R) Direct instruction is OK, but only for slow learners (R) Direct instruction is just part of our normal teaching practices Direct instruction relates to outdated views of learning that no longer apply (R) Direct instruction, as a teaching method, is not consistent with the view that students are active learners who develop personal self-management strategies (R) 11 I would be pleased to discover that my own child was being taught by a teacher using high levels of direct instruction

1 2 3 4 5 6 7 8 9 10

Item statement

TABLE I. Item means, standard deviations, and percent responding in positive attitude direction

486 M. S. Demant & G. C. R. Yates

Teacher Attitudes to Direct Instruction

487

FIG. 1. Teacher attitude (positive directionality) towards direct instruction as a function of gender and teaching experience

3. teacher monitors closely to see if students understand 4. teacher explains things in detail 5. teacher uses a good deal of modelling and demonstrating That is, teachers with positive attitudes generally ticked these items, whereas the teachers with negative attitudes appeared not to recognise that these items form part of the concept of direct instruction. In relation to the checklist tapping into teachers’ awareness of researchers whose names are associated with instructional research, the results were easy to interpret. Of the total of 58 teachers, 46 (79%) did not identify any researcher’s name, 6 (10%) identified one of the seven researchers, and another 6 (10%) identified two or more. No significant relationships were found between this factor and either the attitude or knowledge scores. Identification levels for some of the other names (the foils) were as follows: Cathy Freeman 100%, Jean Piaget 90%, B. F. Skinner 60%, and Carl Rogers 32%.

Discussion In sum, it was found that: 1. The concept of direct instruction did appear to be a suitable target for divergent

488

2.

3.

4.

5.

M. S. Demant & G. C. R. Yates professional opinions, with 19% of respondents (primary school teachers) indicating negative attitudes, and 81% expressing generally positive attitudes. Positive attitudes were found to relate strongly to teachers’ knowledge of the components of the direct instruction construct as articulated by process-product researchers such as Rosenshine and Stevens (1986). There was some variation associated with teacher characteristics, with female and more experienced teachers indicating more positive attitudes. Junior and upper primary teachers, however, reported similar attitude levels. The only attitudinal statement which seemed to attract a high level of disagreement (negative attitude direction) was “Direct instruction is a highly effective teaching method with all students” which had a disagreement level of 53%, an agreement level of 39%, with the other 8% expressing the neutral point on this item. As a group, the respondents appeared unable to identify any researcher whose work has been associated with instructional research.

The results indicate a high level of support for the direct instruction construct, especially in those teachers who appeared to be aware of what the term refers to within contemporary research. Space prevents us from reporting the answers to several open-ended questions on the original questionnaires. But in summary, we were impressed by the sensible and professional attitudes expressed. The notion that experienced practitioners are generally favourable towards direct instruction receives strong support from the current data. The fact that attitudes were less positive in younger and male teachers is not easily interpreted. It is tempting to conclude that as teachers become more experienced and skilful they adopt a balanced and pragmatic attitude towards their instructional practices, and see direct instruction as just another useful and necessary professional tool. The fact that male teachers were less positive than female teachers, however, is a finding we are at a loss to account for. All we can note is that, in the present sample, there appeared to be a cluster of younger male teachers who responded around the midpoint on our attitude scale (neither positive nor negative attitude direction). But whether or not such a finding will generalise beyond the current sample, or to teachers at the high school level, is quite unknown. The finding that teachers with positive attitudes tended to be more aware of the research-related meaning (teacher effects) of the term direct instruction suggests that the negative images associated with direct instruction by some agencies are linked to active misconceptions of what the term means. Slogans such as “chalk and talk”, and “drill and kill” are occasionally promulgated as though they are supposed to represent the same practices as those identified by modern researchers such as Rosenshine and others such as Good and Brophy (1999). Indeed, Baumann (1988) has detailed various myths that have been levelled at the direct instruction construct, and we drew on his list in constructing the present scale, especially the reversed items as shown in Table 1. The present data suggest that teachers in general, and experienced teachers in particular, are not misled or confused by such misconceptions.

Correspondence: Dr Gregory Yates, School of Education, University of South Australia, St Bernards Rd, Magill, SA 5072, South Australia (e-mail [email protected]).

Teacher Attitudes to Direct Instruction

489

REFERENCES Baumann, J. F. (1988). Direct instruction reconsidered. Journal of Reading, 31, 712–717. Chall, J. S. (2000). The academic achievement challenge: What really works in the classroom. New York: Guilford. Cobb, P. (1988). The tension between theories of learning and instruction in mathematics education. Educational Psychologist, 23, 87–103. Delpit, L. D. (1988). The silenced dialogue: Power and pedagogy in educating other people’s children. Harvard Educational Review, 58, 280–298. Good, T. L., & Brophy, J. (1999). Looking in classrooms (8th ed.). Boston: Allyn and Bacon. Speigel, D. L. (1992). Blending whole language and systemic direct instruction. Reading Teacher, 46, 38–44. Rosenshine, B., & Mesister, C. (1995). Direct instruction. In L. W. Anderson (Ed.), International encyclopedia of teaching and teacher education (pp. 143–149). Oxford: Pergamon. Rosenshine, B., & Stevens, R. (1986). Teaching functions. In M.C. Wittrock (Ed.), Handbook of research on teaching (3rd ed., pp. 376–391). New York: Macmillan.