Proceedings paper


Children's Assessments Of Their Own Compositions

Liz Mellor


This paper draws from my own PhD research on assessing pupils' compositions. Within the National Curriculum in England composing is a statutory part of the programmes of study for Music. From September 2000 the revised orders for music introduce level descriptors to be used by teachers to best fit a pupil's performance at the end of Key Stage 1 (age 7), Key Stage 2 (age 11) and Key Stage 3 (age 14).

From the original implementation of the Music National Curriculum, to subsequent revisions in 1992, followed by the Dearing revision (1995) and the recent simplification (1998), the expectations at each of the three Key Stages have narrowed. This is presented in the form of a sequential progression through the elements of music (pitch, duration, dynamics, tempo, timbre, texture and structure). Thus :

 Key Stage 1

pupils should be 'recognising the musical elements'

Key Stage 2

pupils should be 'distinguishing the musical elements'

Key Stage 3

pupils should be 'discriminating within and between the musical elements'


In 'Teaching Music in the National Curriculum' Pratt and Stephens (1995) presented this model in the form of a table to indicate progression within each element. According to this, it follows that : pupils would first talk about loud, quiet and silence before going on to talk about gradating levels of volume, and before progressing to recognise subtle differences in volume. This implies that musical conceptual understanding is developed though an increasingly discriminatory vocabulary based on the principle of quantitative addition. As Swanwick (1996) argues, most of the Key Stage statements in the Music National Curriculum document are essentially quantitative in character rather than qualitative; he urges 'we need to have criterion statements to pick up these qualitative shifts' (p. 34).

The theoretical framework draws from several research projects which have tried to map out these qualitative shifts. For example, Swanwick and Tillman's model (1986) was adapted by Swanwick to suggest a basis for establishing criteria for assessing composition and more recently for assessing performance and listening (Swanwick, 1988, 1996, Hentschke, 1993). Other researchers such as Flynn and Pratt (1995) use a 'bottom-up' approach. This type of approach sought to make explicit the criteria which the teachers identified in making such qualitative shifts.The DELTA Project (Development of Learning and Teaching in the Arts) conducted by Hargreaves, Galton and Robinson (1996) devised a methodology which claimed to make explicit the implicit criteria which teachers use to make judgements about children's products. The findings for music made ground in developing a language of assessment. Hargreaves et al. (1996) provided a five phase model which incorporated domain specific as well as general cognitive aesthetic developments. The researchers reported the developments in terms of phases so as not to be confused with Piagetian 'stages'. The five phases are denoted as sensorimotor , figural, schematic, rule systems and professional. The model draws from current domain specific research and acknowledges its somewhat sketchy form at the stage of writing as reflecting the parsity of research in this area. Nevertheless, it provides some interesting insights which draw together psychological research into aesthetic development. Rather than attempting to draw out level descriptors Hargreaves and Galton's model sketches phases of development which invite further research and 'real-world' application.

This paper draws from that part of my earlier research which set out to exemplify a 'real world' application. It was also set in the climate of the raised profile of literacy across the curriculum taking into account the reports on the use of Language within the Common requirements of the National Curriculum (SCAA, 1997) and their attempt to provide a way forward for the role of language in music education:

Teachers are encouraging the use of technical vocabulary with greater confidence but more help is requested with regard to the musical vocabulary which should be taught at each key stage. (SCAA 1998, Section 11)

Teachers in all key stages need guidance on subject knowledge and how this knowledge can be integrated in practical work including the development of aesthetic awareness and musical vocabulary. (SCAA, 1998, Summary)

One objective of the research was to investigate how children used the language of the musical elements as defined by the English National Curriculum as pitch, duration, dynamics, tempo, timbre, texture and structure. I was interested to see how the terms were used by the children and to what extent these revealed qualitative shifts in their conceptual understanding of their own and their peers' compositions.


The research included 154 children, 78 girls and 76 boys aged 9-13 years. The sample was taken from Years 5-6 in the Upper Primary School (Upper Key Stage 2: ages 9-11) and Years 7-8 of the Secondary School (Key Stage 3; ages 11-13). The research task was designed as part of the children's curriculum music sessions led by the teacher (new in post) who was also the researcher. The children had been asked to compose 'what they thought made a good tune'. The design took account of the pilot study research in the following ways.

First, the composing task was presented in an open-ended way. To avoid influencing the listening responses the task was not directed in a series of sequential stages. Second, the starting point was a melodic composition; as such, it did not have an extra-musical referent. Third, the composing activity was organised on an individual basis and not in groups, thereby allowing each composition to be identified with each pupil's individual response. Fourth, the school was fortunate to have sufficient keyboards for the children to share one between two. Although there may have been some pair-work influence the children used separate headphones and were asked to work individually on their own tunes. Keyboards were used as a means of 'controlling' the sound sources used, so that the children's subsequent responses were not limited to a simple recognition of the instrument. They also proved to be a highly motivating sound source across the sample age range. For the purposes of the research the pupils were asked to choose their own sound from the sound bank but not to use a rhythm or beat accompaniment. Note names and/or staff notation could be used as a means to map out the tune for performance but this was not obligatory. The children worked on their tunes in 3 x 50 minute lessons and performed and recorded their compositions onto audio tape.

The design took into account that for many children composing was a new experience. Some had more experience of playing an electronic keyboard than others. The research also acknowledged that some children had piano skills and that this would influence the musical outcome. However, it was considered that the task was equally accessible for all children, open-ended enough to allow individual approaches and age appropriate (in both the instruments used and also the type of task). Equally, the task fulfilled the requirements of the Music National Curriculum and was presented in such a way as to encourage pupil ownership of the learning which was synonymous with the school's philosophy of education.

In the final week the children were invited to appraise their compositions. For research purposes the design investigated the childrens' listening responses to their own composition and to those of their peers. In practice the children listened to the recording of their classes' compositions and, in the pause in between each piece, they gave a mark out of 10 and wrote a reason for their choice on a given pro forma. The results were collated and used as a basis for both quantitative and qualitative analysis.

Data Analysis

In order to analyse the data a coding scheme was developed to categorise the content of written listening responses. An initial survey of the data produced 22 categories of response. These were subsequently reduced to five broad categories as follows :

1. Musical Elements

Responses in this category refer to the elements of music as defined in the Music National Curriculum. They include references to: Pitch, Duration, Dynamics, Tempo, Timbre, Texture, Structure. Responses in this category might include for example, 'it was loud', 'it was short', 'the notes went up and down', 'it had a hollow sound', 'it repeated'.

2. Style

In this category are responses which make stylistic references, for example, 'it sounds classical', 'it sounds like Jazz', 'it sounded Japanese'.

3. Mood

Responses in this category indicate an affective response to the music, for example, 'it made me feel happy', 'it was depressing', 'it was spooky'.

4. Evaluation of Composition

Responses in this category demonstrate an evaluative statement of the composition itself, for example, 'it was good', 'it was well put together'.

5. Evaluation of Performance

Responses in this category refer to an awareness of the qualities of the performance, for example, 'he missed a note', 'it was played well' .

An independent rater and myself performed a reliability study which categorised a sample of the responses. The results were correlated using statistical measures and showed that the coding scheme allowed a satisfactory level of agreement and that further analysis was justified.


For the purposes of this paper I shall present a summarised version of the

qualitative results. The initial analysis involved mapping the responses into the five categories described above as Musical Elements, Style, Mood, Evaluation of Composition and Evaluation of Performance. I shall discuss each category in turn focusing on the language used by the children. The extracts from the children's responses were chosen because they were representative of particular types of response.

Musical Elements

Within this category the responses were subdivided into a further 7 subcategories which corresponded to the musical elements within the Music National Curriculum (DFE, 1995). I shall focus on each in turn.


In the Music National Curriculum pitch is described from Key Stage 1-3 as :

(KS 1) high /low

(KS 2) gradations of pitch e.g. sliding up/down, moving by step/leap, names for pitch

(KS 3) various scales and modes e.g. major, minor

The children's responses revealed a range of ways of talking about pitch in relation to their tunes showing a greater degree of aesthetic differentation to the schema presented in the Music National Curriculum. To summarise, responses which refer to pitch show:


In the Music National Curriculum duration is described from Key Stage 1-3 as :

(KS 1) long/short; pulse or beat; rhythm

(KS 2) groups of beats, e.g. in 2s, 3s, 4s, 5s; rhythm

(KS 3) syncopation, rhythm

The responses divided between those which focused on the qualities of the duration as beat or rhythm and those which focused on the duration of the tune as a whole. The children's responses which focused on duration/beat-rhythm were considerably less differentiated than those within the category of pitch..To some extent some of the responses might have been made in relation to the the quality of performance as much as for the rhythmic qualities of the tune. No explicit responses demonstrated an understanding of groups of beats per se.To summarise, responses which refer to duration/ beat-rhythm show:

Apart from one response, which comments on the length of the notes ('[it's] good how it is staccato'), most of the children's responses which use the words long and short refer to duration as the length of a the tune as a whole. In this way duration is linked to the element of structure. To summarise, responses which refer to duration/long-short show:


In the Music National Curriculum dynamics are described within Key Stage 1-3 as :

(KS 1) loud, quiet, silence

(KS 2) different levels of volume, accent

(KS 3) subtle differences in volume, e.g. balance of different parts

Not all the keyboards were touch sensitive and so the volume was controlled at source rather than through touch. The most marked difference in this sub-category was that the Key Stage 2 children made far more references to dynamics than the Key Stage 3 children. From this initial analysis there is evidence to suggest that the girls produced more responses which showed a preference for quiet music and boys produced more responses which favoured loud music, especially in upper Key Stage 2. To summarise, responses which refer to dynamics show:


In the Music National Curriculum tempo is described from Key Stage 1-3 as :

(KS 1) fast, slow

(KS 2) different speeds, e.g. lively/calm, slower/faster than;

(KS 3) subtle differences in speed, e.g. rubato

Far fewer children produced references to tempo at Key Stage 2 and girls referred to this more than boys. However at Key Stage 3, far more young people produced references to tempo and boys referred to this more than girls. The initial analysis also shows that the boys' responses showed a preference faster music. To summarise, responses which refer to tempo indicate:



In the Music National Curriculum timbre is described from Key Stage 1-3 as :

(KS 1) quality of sound, e.g. tinkling, rattling, smooth, ringing

(KS 2) different qualities, e.g. harsh, mellow, hollow, bright

(KS 3) different ways timbre is changed, e.g. by mute, bowing/plucking, electronically; different qualities, e.g. vocal and instrumental tone colour

To summarise, responses which refer to timbre express :



In the Music National Curriculum texture is described from Key Stage 1-3 as :

(KS 1) several sounds played or sung at the same time/one sound on its own

(KS 2) different ways sounds are put together e.g. rhythm on rhythm; melody and accompaniment; parts that weave, blocks of sounds, chords.

(KS 3) density and transparency of instrumentation; polyphony and harmony

There were very few responses in terms of texture and this can be accounted for by the nature of the composition task. This was essentially a linear melodic construction and did not require more than one part at once. Some children used chords to accompany their melodies and some responses reflect this e.g. 'long, with nice chords', 'the chords go well together'.


In the Music National Curriculum structure is described from Key Stage 1-3 as :

(KS 1) different sections, e.g. beginning middle end, repetition e.g. repeated patterns, melody, rhythm;

(KS 2) different ways sounds are organised in simple forms, e.g. question and answer, round, phrase, repetition, ostinato (a musical pattern that is repeated many times), melody;

(KS 3) forms based on single ideas e.g. riff, forms based on alternating ideas e.g. rondo, ternary, forms based on developmental ideas e.g. variation, improvisation.

The responses showed that the pupils perceived structure in a number of ways.

To summarise, responses which refer to structure show:


Children's style sensitivity is represented in a number of ways. To summarise, responses in this category refer to:

Children who responded in this way have picked out a quality in the sound, such as the use of the Indian sitar in the sound bank, or a musical feature, such as the intervallic pitch relationships in the 'Egyptian tune', or the syncopated rhythm of the 'Caribbean tune'. As they do not yet have the vocabulary to describe the specific musical features, they use stereotypes.

Whereas at Key Stage 2 style responses were dominated by references to film, video and TV, Key Stage 3 pupils' perception of style related to personal experience, preference and identity.


Children responded in this category in a number of ways. To summarise, responses in this category include:

Evaluation of Composition

To summarise, responses in this category include :

As above, many children justified their preference by valuing one or more aspects of the tune in the categories of Musical Elements and Style.

Interestingly the focus of the responses showed that the younger children were more likely to express themselves in terms of whether the piece was copied whereas the older Key Stage 3 children were more concerned with the quality of originality, difference, imagination and creativity.

Evaluation of Performance

Children produced different types of response in this category. To summarise, responses in this category refer to :



To a certain extent the responses in each sub-category confirm the way the Music National Curriculum defines the increasing levels of discrimination within each musical element i.e. pitch, duration, dynamics, tempo, timbre, texture and structure across Key Stages 1-3. The results therefore provide verbal evidence of how children listen and appraise music in relation to the English Music National Curriculum .

However, the results also show ways in which the children's musical understanding becomes increasingly differentiated both within each sub-category and between categories of perception, beyond the definition presented within the documentation of the Music National Curriculum. This gives a more detailed picture of how children use language in their responses in each of the subcategories and more particularly shows us what they value. Responses also change across the categories, within the categories and with respect to age and gender, and leads to a fuller picture of the qualitative shifts in the conceptual understanding of music. This is an area for future research.

The results also reveal that responses which show an absence of technical vocabulary may nevertheless communicate a sense of the music. However, some responses use technical statements inappropriately e.g. 'out of tune' was used of the intervallic range within the pitch contour. The conclusion to be drawn is that the use of technical vocabulary may not be evidence of musical understanding.

Another consideration in the analysis is how far the responses were influenced by musical expertise and peer group issues of perceived musical expertise, status, friendship and competition. This is illustrated by examples from the qualitative data which take into account biographical and social observations of the children (Mellor, 1999). For example, the experienced pianist responds with a voice of expertise : 'could have practised more', 'original and good for someone who doesn't play the piano'. The saxophonist with an experience of playing jazz responds using a phraseology common to jazz style e.g. 'doesn't make the most of the rests, needs to sit back on the beat'.

From my experience as teacher and researcher some responses reflect the relative social status of the children within the class. The use of the term high/low status is defined by my observation of how the children interacted and whom they held in esteem amongst their peers. The qualitative analysis therefore reveals that additional factors need to be taken into consideration. The particular value of a teacher/researcher is the ability to analyse internal social hierarchy within classes which produces another level of subjectivity beyond that of gender.

As the full research shows, whilst some responses share characteristic patterns or phases of development, concerned with issues of recognition, conformity, appropriateness, originality and reflection, listening responses show individualised profiles which are mediated by the listening context and the social structure of the group. The question for policy makers and music education research must be how to integrate these observations into the assessment model. Whilst general guidelines may be welcomed, over simplification as presented by the Qualifications Curriculum Authority (2000) might be limiting and misleading. In seeking to 'level out' the types and range of performance that pupils demonstrate I hope we don't inadvertently 'level out' the richness of this inquiry which is still a largely uncharted territory.


Department of Education and Science (1991) Music for ages 5-14. London: HMSO.

Department of Education and Science (1992) Music in the National Curriculum. London. HMSO.

Department for Education (1995) Music in the National Curriculum, London: HMSO

Flynn, P. and Pratt, G. (1995) Developing an understanding of appraising music with practising primary teachers. British Journal of Music Education, 12, 127-158.

Hargreaves, D.J., Galton, M.J. and Robinson, S. (1996) Teachers' assessments of children's classroom work in the creative arts. Education Research, 38, 2, 199-211.

Hentschke L. (1993) Musical development; testing a model in the audience-listening setting. Unpublished PhD Thesis. Institute of Education: University of London.

Mellor L. (1999) The Language of Self- Assessment: Towards aesthetic understanding in music. in E. Bearne (Ed.) Use of Language across the Secondary Curriculum..London: Routledge.

Pratt, G. and Stephens, J. (1995) (Ed.) Teaching Music in the National Curriculum. National Curriculum Music Working Group. Oxford: Heinemann.

Qualifications and Curriculum Authority (1998) Breadth and Balance in the National Curriculum

School Curriculum Assessment Authority (1997) Music and the Use of Language at Key Stage 3, London: SCAA.

Swanwick, K and Tillman, J (1986) The sequence of musical development. British Journal of Music Education 3, 305-39.

Swanwick, K. (1996) Music before the National Curriculum. In G. Spruce (Ed.), Teaching Music. London: Routledge.


 Back to index