Primary teachers' understanding in science and its impact in the classroom
Paper presented at the British Educational Research Association's Annual Conference, Lancaster, September 1996
Wynne Harlen, Scottish Council for Research in Education
Introduction
Research carried out over a period of two years into primary teachers' understanding of concepts in science and technology provided information about teachers' confidence in teaching science, their understanding of a range of key ideas and their background of science in their own education. The early part of the paper describes how these data were collected, noting that the method of interviewing used to explore teachers' understanding served as an opportunity for the advancement of their understanding. The relationships among the variable of confidence, understanding and background in science are then explored.
Although teachers expressed low confidence in teaching science, which was linked to lack of understanding of scientific ideas, they also claimed not to find great difficulty in using certain teaching skills required for science, including those which would appear to demand content knowledge. This apparent anomaly is discussed and linked to the strategies used by teachers to cope with low confidence and understanding. The paper ends by considering briefly the role of content knowledge in teaching science.
The research
The research on which this paper draws was a two year project, funded by the Scottish Office Education Department, and conducted from 1993-5. Different kinds of data were gathered in four phases of the project. In Phase 1, 514 primary teachers responded to a questionnaire about their background in science and their confidence in various aspects of teaching science and technology. In Phase 2, 57 teachers, a subset of those in Phase 1, were interviewed about their understanding of ideas in science and technology. In Phase 3, 33 teachers kept notes about their teaching during a period of ten weeks and were interviewed about this work by telephone. In Phase 4, 30 members of staff in teacher education institutions and some local authority advisers were interviewed about the initial and inservice education in science of primary teachers.
The survey conducted in Phase 1 used the sample of schools taking part in the 1993 Assessment of Achievement Programme's (AAP) survey of science performance of pupils. This ensured not only a representative sample of schools in Scotland but a response rate (83%) which is unusually high for a questionnaire survey, giving particularly dependable findings.
The data reported here derive from the first three phases of the project. Phase four confirmed that, at the time of the research, the science courses in initial teacher training were limited both in duration and in coverage. There was an increasing attention content but a reluctance to move too far from the process focus.
It should be noted that, when the research began in 1993, the final draft of the National Guidelines: Environmental Studies 5-14 (SOED, 1993), of which science and technology are two of five components, had only just reached schools. Attention in almost all schools was at that time focused on implementing the guidelines for English language and mathematics. Thus the research reports on the situation before the impact of the guidelines was felt in schools and before the range of materials from the Scottish CCC and from local authorities and an extended provision for inservice became available. although the research project referred to both science and technology, this paper is concerned only with the part relating to science.
The teachers' experience and science background
Of the 514 respondents in the survey sample, 8% were male and 92% female. This agrees closely with the distribution by sex of all teachers in education authority primary schools in Scotland. 248 (48%) were teachers of P3 and/or P4; 259 (50%) were teachers of P6 and/or P7; seven teachers taught some other class or combination of classes or did not tell us what class they taught. Thirteen (3%) had responsibility within their school for science and four for technology. 72% of teachers in the sample had been qualified for 12 years or more; 28% had been qualified for less than 12 years. 65% had been teaching for 12 years or more; 35% had taught for less than 12 years. 21% of the teachers had a degree qualification Ð either a BEd or some other degree plus PGCE. The remainder were diplomates of a college of education.
Twenty teachers only (i.e. 4% of the total) had a degree containing a science subject; in 18 of these 20 cases the subject was a biological science. Only two teachers in the sample had a degree which included a pass in a physical science subject. A further 88 teachers had at least one Higher grade pass in a science subject. A further 90 teachers claimed at least one Ordinary ('O') Grade science pass. Three hundred and fourteen teachers (61%) said they had no science qualifications.
The gender, teaching experience and science background of the teachers taking part in Phases 1 and 2 are summarised in Table 1
Table 1The teachers in the phase 1 ( N=514) and phase 2 samples (N=57)
Characteristic / Number of teachers in Phase 2 / Percentage of teachers in Phase 2 / Percentage in Phase 1 sample(N=514)
(a)Sex / Male / 4 / 7% / 8%
Female / 53 / 93% / 92%
(b)Teaching experience / 12 years or more / 39 / 68% / 72%
Less than 12 years / 18 / 32% / 28%
(c)Science background / No science / 38 / 67% / 61%
Some science / 19 / 33% / 39%
Teachers' confidence in relation to science
In the survey the teachers were asked to respond to the following question in relation to each of a series of statements selected from the National Guidelines indicating the content and contexts through which upper primary pupils should develop their understanding in science:
'How confident do you feel that you have the knowledge needed to help pupils develop understanding of each of the following kinds, quoted from the 5-14 Guidelines for Environmental Studies? Please rate them 1-4 using the following criteria
1 Fully confident in my knowledge and skills
2 Confident with a little guidance from others
3 I can manage but depend on advice from others
4 I need help to develop my knowledge and skills in this area'
For the purposes of analysis the statements were sorted into groups relating to the 'attainment outcomes' of understanding living things and the processes of life, understanding energy and forces and understanding Earth and space. Within each group some statements are taken from those designated in the Guidelines as being for pupils in P4 - P6 (years 4 - 6) and some for P7 to S2 (years 7 - 9). As an example of the detailed findings, Table 2 gives the results for the sample of statements relating to understanding energy and forces.
Table 2 TeachersÕ confidence on aspects of ÔEnergy and ForcesÕ: percentages of teachers responding in each of the four categories (N=514)
Stage / Cat 1 / Cat 2 / Cat 3 / Cat 4 / ConfidenceIndex
P4/6 / Mirrors and reflections, including curved mirrors / 15 / 36 / 34 / 16 / 252
P4/6 / Sound from a variety of vibrating sources eg tuning fork, violin string, radio and TV speaker / 15 / 32 / 35 / 18 / 244
P4/6 / Force of gravity / 13 / 33 / 33 / 22 / 239
P4/6 / Construction of battery-operated circuits to operate eg bells, buzzers, lamps, warning lights. / 17 / 29 / 29 / 25 / 238
P4/6 / EarthÕs magnetic field and the compass / 13 / 32 / 35 / 20 / 238
P7/S2 / Renewable and non-renewable sources of energy / 12 / 33 / 32 / 23 / 234
P7/S2 / Potential and kinetic energy / 8 / 21 / 28 / 43 / 194
AverageÐover five P4/P6 items
Ðover two P7/S2 items / 242
214
The confidence index gives a summary for each statement that takes account of the judgements across the range of categories. It is formed by multiplying the percentage of responses in category 1 by 4, in category 2 by 3, in category 3 by 2 and in category 4 by 1. The maximum confidence index for any item is thus 400, the minimum 100 and the median is 250. Using these indices facilitates comparison across the different content outcomes, as in Table 3.
Table 3Summary of percentage responses for science content outcomes across categories (N=514)
Cat 1 / Cat 2 / Cat 3 / Cat 4 / AverageConfidence
Index
Understanding Living Things and the Processes of Life / 27 / 38 / 24 / 11 / 282
Understanding Energy and Forces / 13 / 31 / 32 / 24 / 234
Understanding Earth and Space / 27 / 38 / 26 / 9 / 284
Teachers appear to be more confident about their knowledge of biological topics and Earth and space and less confident about energy and forces. In response to similar questions about confidence in respect of developing pupils' process skills, the average confidence index was 282 and for all items the value was above the median of 250.
These are not unexpected findings and corroborate what was found by Bennett, Wragg and Carre just before the introduction of the national Curriculum in Science in England (Carre and Carter, 1990; Bennett et al, 1992). However there was an apparent conflict between teachers' lack of confidence reported in these tables and the responses they gave when asked about how difficult they found various teaching skills required in science. In table 4 the meaning of the categories is as follows:
Category 1=Not at all difficult
Category 2=Sometimes a little difficult
Category 3=Usually quite difficult
Category 4=Very difficult
Table 4 TeachersÕ estimates of difficulty of certain skills in teaching science (Items listed in order of increasing difficulty) (N=514)
Cat 1 / Cat 2 / Cat 3 / Cat 4 / DifficultyIndex
Introducing a new topic / 58 / 39 / 3 / 0 / 145
Ensuring the equal interest and participation of boys and girls / 50 / 39 / 10 / 1 / 162
Explaining ideas to pupils / 41 / 54 / 5 / 1 / 168
Responding to pupilsÕ ideas about content / 41 / 53 / 6 / 1 / 169
Using questioning skills to stimulate pupilsÕ thinking / 39 / 50 / 10 / 1 / 173
Matching activities to pupilsÕ development and needs / 23 / 55 / 18 / 3 / 199
Deciding concepts to be developed in an activity / 17 / 64 / 16 / 3 / 205
Deciding process skills to be developed in an activity / 17 / 59 / 20 / 4 / 211
Organising and supporting practical work / 17 / 57 / 23 / 3 / 212
Record keeping / 15 / 43 / 32 / 10 / 237
Obtaining and maintaining equipment / 11 / 43 / 36 / 10 / 245
Continuous (diagnostic) assessment in relation to concept development / 8 / 45 / 35 / 12 / 251
Continuous (diagnostic) assessment in relation to process skills / 7 / 46 / 36 / 11 / 251
Note: The ÕDifficulty IndexÕ is calculated in a similar way to the ÔConfidence IndexÕ. The maximum possible difficulty index is 400; the minimum is 100.
Many of these items refer to skills dependent on pedagogic content knowledge rather than content knowledge (Shulman, 1987). However, since pedagogic content knowledge depends, inter alia, on content knowledge, the low level of difficulty for some items, particularly 'explaining ideas to pupils' and 'responding to pupils' ideas about content' is at first surprising. It appears that teachers have less difficulty in coping with these demands than might be predicted from their perceived confidence. The important question that must be asked, of course, is how are they coping? Although this research was not able to observe teachers in action, their own accounts, given in interviews in Phases 2 and 3 of the project indicated that in such situations teachers were using a combination of their considerable general pedagogic skills and strategies for avoiding them in the first place. These are reported later (page 10).
Teachers' understanding of scientific ideas
The evidence for this came from Phase 2 of the project. The method used was devised to put teachers at ease and to avoid at all costs, the impression that they were being 'tested' and perhaps judged as inadequate. In brief, a series of 'events' was devised, each relating to one of the selection of statements used in the Phase 1 survey. For each event three 'big ideas' were identified and used as a structure in the interviews and as a framework for analysis of the tape recordings of the interviews. The focus of the interview was the event, which was presented either with simple equipment or by means of good coloured photographs. The form that the discussion of this event took led to its being called a 'collaborative explanation'. The aim was to arrive at an explanation for the event or phenomenon which was satisfactory to both the interviewer and the teacher being interviewed. In some cases the teacher provided the explanation and in some cases the interviewer proposed ideas to test out, at all points checking that what was suggested made sense to the teacher in terms of the evidence in front of them and other evidence that could be recalled.
The resulting experience was not an unpleasant one for the teachers and indeed some expressed pleasure at understanding something which they had not understood before. At the same time it gave the interviewer the opportunity to probe the understanding that the teacher had to begin with and the extent to which this changed during the interview. There were six events and each teacher was interviewed about three of these.
The tape-recorded interviews were analysed for evidence of understanding of the big ideas related to each event and for any change in understanding. A list of the events and the big ideas is given in the Appendix. For each idea it was noted: whether understanding was shown without any prompting; whether understanding did not appear to be present initially but was shown by the end of the discussion; or whether there was no evidence of understanding at any point. A category of 'not sufficient evidence' was also used since, although there was specific and trialled sequence to be followed by the interviewer it was not always possible, in a fairly free ranging discussion, to bring the conversation back to points that may have been left incomplete.
The results were used to form a hierarchy of big ideas in terms of their 'ease of understanding'. The most easily understandable ideas included:
Water exists as solid, liquid or gas
Current flow needs a circuit of suitable materials
Switches make and break the circuit
Bones move at joints because of muscles
and the least easily understood included:
Evolution occurs over time because of the selection of those inherited differences which are helpful for survival
The battery supplies electrical energy which is changed in the bulb to heat/light energy
The brain 'assumes' that light has followed a single straight line to the eye even if it has been reflected on the way
An analysis was also made of the misunderstandings that the teachers displayed. It is not the purpose to present these here ( a full account in given in Harlen et al, 1995) but two things should be noted. First, like the ideas that children often hold, the teachers' non-scientific ideas were the result of rational thinking but based on links or analogies or use of words that were inappropriate. Second they were in many cases corrected by the teachers themselves once they had chance to consider relevant evidence or alternative ideas. Indeed, on average 36% of the big ideas were held by teachers at the start of the interviews, whilst there was evidence of understanding being developed or revealed during the interview for 44% of the big ideas and for 19% there was no understanding shown (1% insufficient evidence).
As well as looking at the understandability of the big ideas the interviews gave a rough measure of the understanding of each teacher. Thus for 55 teachers who completed both Phase 1 and Phase 2 of the research (interviews were not completed in two cases) there was information about their background in science, their confidence in teaching and their understanding. The relationships between these variables could then be explored.
Science background and understanding
Using information about extent of understanding at the end of the interview, teachers were identified as being in the top one-third or bottom one-third of the group on the basis of understanding. The characteristics of the teachers in the top and bottom one-third were then identified. Table 5 sets out the findings for length of experience and science background. Percentage figures for the Phase 1 sample are given for comparison.
Table 5The experience and science background of teachers in the top and bottom thirds with respect to understanding
Top thirdN=18 / Bottom third
N=18 / Phase 1
N=514
Experience
12 years or more / 12(67%) / 13 (72%) / (65%)
Less than 12 years / 6(33%) / 5(28%) / (35%)
Science background
Some / 10(56%) / 0(0%) / (39%)
None / 8(44%) / 18(100%) / (61%)
Table 5 shows that the top 18 and the bottom 18 teachers are not significantly different with respect to length of teaching experience Ð and not significantly different from the sample of teachers as a whole. Those in the top third with respect to science understanding are more likely to have some science in their background than not, whereas there is no-one in the bottom third who has some science in their background. Eight of 18 in the top third for science understanding have no science in their background. That these teachers have, by their own efforts or with assistance, gained adequate or good understanding of those ideas that are to be developed in pupils was confirmed by interview data where , for example, some teachers mentioned that before teaching any topic they would 'read it up' beforehand. Other teachers, however, lacked the confidence, or perhaps the knowledge of where to 'look it up' although they were ready to do this in other areas:
In other areas, with a difficult question I always have some source and I can tell a child 'Give me a few minutes or till tomorrow' and I can come up with that answer, but in science I genuinely cannot say that.
Teachers who had only biology in their background were investigated separately; not surprisingly their understanding of biological ideas was better than their understanding of physical science ideas.
There is nothing very surprising about understanding of these science ideas is on the whole better amongst those teachers who have science in their own background. However it is relevant to recall the criterion used for having some science - just one science subject at 'O' or 'S' grade (equivalent of GCSE) level. Was this enough to give confidence? This is the next question to be addressed.
Confidence, understanding and background science
The sample, it was noted at the start of this paper, contained only 8% male teachers, yet there was a statistically significant differences in confidence related to gender. Male teachers were more confident in relation to 17 out of the 19 items in the survey relating to developing children's understanding and in the remaining two there was no significant difference. Thus the presence of even a small number of male teachers was introducing a gender difference which could interfere when looking at the effect of other variables. Thus in dealing with the smaller number of teachers for whom there was the full range of data, it was considered more helpful to eliminate the gender element in looking for the effect of other variables. The two men for whom complete data were collected had some science in their background ('O' grade) and had high confidence and high understanding. Table 6 gives the findings for 34 female teachers who took part in phases 1 and 2.