role of task variability and home contextual factors in the academic performance and task motivation of African American elementary school children, The

Journal of Negro Education, The, Winter 2001 by Bailey, Caryn T, Boykin, A Wade

The score on the PPM is used to compute two different expressions of variability preference. One method, the pathway mean score, determines the mean variability preference across the five trials, whereby the mean of the five trial scores is obtained. The second method, the pathway variance score, determines the change in variability preference across the five trials, whereby the variance among the five trial scores is computed.

Stimulus Materials

Four academic task-types were constructed for this study. They were a spelling, vocabulary, math, and picture-sequencing task. These specific task-types were chosen because they assess the kind of cognitive problem solving skills discerned in actual classroom settings and on standardized achievement assessments. Each academic task-type is described below:

1. Spelling. The children were administered a short story (approximately five paragraphs in length) containing approximately 27-37 misspelled words. The children were given 60 seconds to read the story and circle the misspelled words found in the text.

2. Vocabulary. The children were presented a worksheet containing 15 fill-in-the-blank sentences. Four multiple choice vocabulary word options were provided for each sentence. Children were allowed 60 seconds to read all the sentences and select the vocabulary words that best completed the sentences.

3. Math. A worksheet containing 24 addition operations was given to the children. In a period of 60 seconds, the children were asked to solve as many problems as possible.

4. Picture Sequencing. This task contained 15 picture scenes of an object (e.g., a television, a house, a clock, a turkey, etc.). Each picture scene varied in the amount of illustration detail. The children were given 60 seconds to identify the correct sequencing of the picture scenes, from the scene having the least amount of illustration detail to the completed picture scene having the greatest amount of detail.

Ten exemplars of each of the four academic task-types were developed (i.e., 40 tasks in all). The 40 exemplars were then divided in half and assigned to one of two task sets. Task assignment to a set was conducted in a manner to ensure that a given task exemplar in one set was matched as closely as possible with its counterpart in the other set on the basis of their surface characteristics (i.e., in terms of the number of misspelled words contained in a spelling task or the type of object contained in a picture-sequencing task). A given task exemplar remained with the assigned set for the duration of the experimentation.

Task Conditions

The two task sets were presented to children in two different contexts: the low variability context and the high variability context. In the low variability context, children received 20 tasks (five exemplars of each of the four academic task-types) in a blocked sequence format according to task-type. That is, they were presented all five tasks of one type, followed by all five tasks of a second type, etc. In the high variability context, children received 20 tasks in a random sequence, without regard for task-type. The tasks were counterbalanced for presentation in the low variability context using a Latin square procedure. Four different orders of the task-types were generated. In the high variability context, four different random strings of the 20 tasks were generated with the following qualifications: in every block of five tasks each task-type appeared at least once, no more than two tasks of the same type appeared in succession, and there could not be more than two successive pairs in which both tasks were of the same type. Counterbalancing was also performed to eliminate systematic bias attributed to task variability context order. Consequently, half of the children completed the low variability context first, while the other half completed the high variability context first.