¹Texas Woman's University and ²University of North Texas

The objective of making children aware of the steady beat and designing instructions to help students feel the "heart beat" of the music is basic to many methods of music education (Abramson, Chosky, et al., 1986). Abramson et al., note that teachers often find it difficult to design instruction to help students with basic skills in rhythm and jokingly describe the American view of rhythm as "you either got rhythm or you ain't" (Caldwell, 1992). Standard methods of teaching rhythm to young children ignore the individual differences in rhythmic abilities, making such methods often too demanding or not challenging enough. Identifying the children's rhythmic abilities before introducing them to an instructional program designed to develop rhythmic skills will allow teachers to tailor the instruction to the individual child's needs. Such a tailored program should result in the child experiencing more frequent successes and less frustration.

In order to accurately evaluate an individual child's performance on the test, future research will use the test to develop normative scales for different age groups of children. Using these norms teachers will be able to determine whether a child performs better or worse than average on this test. Using the normative data to identify rhythmic abilities could help the teacher determine instructional strategies, screen students for percussion training, or identify particularly talented children.

The purpose of this research was to design a computerized assessment tool of rhythm skills. Two rhythmic skills–synchronization and rhythm pattern imitation & timing–were identified to be used in this test. Synchronization, or the ability to coordinate one's music-making rhythmically with another musician, is a "skill that is distinct from the mere perception, production, or reproduction of rhythms" (Obeler & Fein, 1988, p. 147). The skill of rhythm pattern imitation & timing involves the ability to remember and repeat a rhythmic pattern. Some children are able to advance in their skill on a musical instrument, but have difficulty with ensemble playing. Other children play well in an ensemble, but play imprecisely alone (Obeler & Fein, 1988).

Weikart designed an instrument, The Rhythmic Competency Analysis Test, to assess the synchronization skills of first through sixth grade children (Weikart, 1989). Flohr adapted the Rhythmic Competency Analysis Test and generated the Rhythmic Performance Test (Flohr, 1996; 1993, 1992, 1986). The Rhythmic Performance Test - Revised (RPT-R) presented here extends the testing to include synchronization with an external beat and the tapping of rhythm patterns using a computer.

The RPT-R is a two-part computer generated test developed on the MS-Windows platform. Part I contains five items. Each item is a MIDI recording of an Irish folk song, "Mountain Top." The tempi used for the items in part I were 110 beats per minute (bpm), 120 bpm, 130 bpm, 140 bpm, and 150 bpm.

Part II of the test contains 20 rhythm patterns that are to be imitated by the subject. The task involves listening to a rhythm pattern and remembering it long enough to repeat the pattern by tapping one of the keys on the computer keyboard. The 20 patterns contained lengths of 1-3 measures, meters of 2/4, 3/4, 4/4, and 6/8, tempi of 60 bpm, 90 bpm, and 130 bpm, and durations of eighth notes, quarter notes, and quarter rests. Short rhythm patterns were repeated in the two, three, and four measureitems to help with memorization of the patterns (Jones, 1976; Bregman,1990).

The manner in which an individual performs a rhythmic pattern can be examined from different perspectives. Dalcroze categorized rhythmic performance into three types. Arhythmic performance is spastic and off beat. Errhythmic performance has all the notes in the right place but is dull and boring with no nuance of weight, motion and time. Eurhythmic performance balances motion and rhythm and solves the problem of how to make music move so that the audience is moved. To perform in a eurhythmic way the musician must be able to perform in an errhythmic, synchronized way. The RPT-R is designed to differentiate between arhythmic and errhythmic performance, but cannot identify eurhythmic performance.

Thirty-two children (mean age = 8.26 years; SD = 1.97) participated in a reliability study of the RPT-R test and performed the test twice in a 2 week period. The parents signed an informed consent form, and data was collected in various elementary school programs in the Denton, TX area. The computer collected all of the data. Before testing started, demographic data was entered on a form on the computer. The computer program placed this data in the participant's data file. After completion of this form, testing started with the five items of the first part of the test that required the child to synchronize with the music being played by the computer. While the child played along with "Mountain Top" the computer recorded the child's deviations from the criterion rhythms in milliseconds. If the child responded too early the computer recorded a negative score. If the response was too late the computer recorded a positive score.

These raw scores have been referred to as Constant Error (CE) in motor behavior and reflect a person's bias in performance (e.g., Schutz, 1977; Schutz & Roy, 1973). If a person's average CE is negative, one may conclude that this individual tends to respond before the criterion beat when tapping along with a rhythm. If an individual obtains a positive average CE, this person tends to respond too late and after the criterion beat. However, negative and positive scores can cancel each other out when averaging and could lead one to believe that the performance of an item was exactly on target and error-free. Consequently, it is important to also report an error term that is not affected by scores of opposing signs. Absolute Error (AE) has been used for this purpose. AE is the absolute value of the CE scores (one drops the sign of the CE score before averaging). Schutz and Roy (1978) have shown that AE represents overall error of the performance. As stated earlier, performance can be judged on its bias, but it should also be examined for its variability. The standard deviation of the CE scores reflects the variability (or consistency) in the performance. Schutz and Roy (1978) provided mathematical proof that AE was a composite of CE and VE and thus incorporates the bias in a performance as well as the variability. The AE score is the most straightforward to interpret: a smaller AE score indicates better performance. The computer program calculates AE and VE scores using the CE scores. It also calculates a Percent AE score. Percent AE was calculated by dividing the AE for each beat within an item by the total criterion duration for that item and multiplied by 100. Percent AE is a unit-less variable and allowed for the comparison among different tempi. The results are presented in the table below. Note that all numbers are in milliseconds except Percent AE.

A factor analysis confirmed that the two parts of the test did indeed test different abilities. It produced two factors. The first part of the test loaded high on Factor 2, but not on Factor 1, while the second part of the test loaded high on Factor 1, but not on Factor 2.

The RPT-R has several implications for music education. First, it may be used for screening purposes to determine whether children and adults have rhythmic abilities. Second, the test may be able to show that an individual may be more effective as an ensemble player or a soloist through it synchronizing and imitation parts. Third, the test may be used in a test-retest situation to determine whether a teaching program had a positive impact on the children's rhythmic abilities. These questions will be addressed in future research.