The acquisition of formal scientific reasoning by physical science pupils in standards eight and nine
Chandler, Heather Anne
The experimental training study was aimed at inducing combinatorial reasoning in adolescents. The training was Piagetian in nature but functioned simultaneously as an operationally def'ined Ausubelian advance organizer for novel learning material in chemistry. The sample consisted of 161 pupils from an English-speaking co-educalional high school in Johannesburg. The study employed a pretest-posttest control group design. The pretest consisted of Piaget's first chemical experiment. A technical procedure for trouble-free administration of this task is described. A detailed analysis of the structure of the task is also presented. Training presented the conceptual framework, embodying, inter alia, a systematic approach, and three analogous problems. Pupils were trained individually. Criterion tasks consisted of the pretest in disguised form and a combinatorial problem involving switches which controlled an electric train set. The experimental group obtained significantly higher scores than the control group on both tasks. Prior to training, observed developmental levels of the experimental group were 24% concrete. 41% transitional, 24% early formal and 11% late formal. After training, these levels had improved to 1% concrete, 14% transitional, 80% early formal and 5% late formal. Training was detrimental to most late formal subjects. Specific transfer was satisfactory since criterion tasks differed from the training task. A delayed test on the train evaluation task, three weeks after reinforcement of training, showed no significant difference from the immediate test. The intrinsic effects of age, IQ and sex on the criterion tasks were not usually significant but older subjects of higher IQ tended to achieve better. However, higher IQ Standard Nine boys derived less benefit from training than other subjects. Pupils received instruction on rates of chemical reactions. The experimental subjects performed significantly better in generating the subsumer consisting of the concept underlying the topic. They also tended to achieve better on the recall of factual knowledge pertaining to this topic, reaching significance for Standard Nine boys. Older subjects of higher IQ tended to obtain higher scores but not significantly so for the sample in general. IQ, rather than training, contributed to performance on issues requiring insight into the subject-matter.
A thesis submitted to the faculty of education. University of the Witwatersrand, Johannesburg in fulfilment of the requirements for the degree of Doctor of Philosophy Johannesburg 1984