This study compares methods to evaluate the outcomes of laboratory instruction in high school chemistry and reports the instruments developed for that purpose.
The objectives evaluated were: the ability of students in basic laboratory skills, ability of pupils in the selection of materials, apparatus and methods; and facts that are outcomes of laboratory instruction. These three objectives were selected from some fourteen general objectives gleaned from the literature pertaining to laboratory chemistry. They were chosen as representing outcomes due solely to laboratory instruction as compared with others that may have been achieved at least in part, by the routine lessons.
The experimental method was to evaluate 72 high school students of chemistry by means of:
1. a practical test of laboratory work designed to conform with the objectives chosen referred to as the criterion test.
2. a group pencil and paper test somewhat parallel to the criterion test.
3. the laboratory notebooks of the students.
4. the teacher's estimates of student progress toward the objectives.
Three classes of chemistry were evaluated in the Spring of 1952. The teacher's estimates were prepared in February from observation of the students at work in the laboratory. The laboratory reports had been marked weekly for six months prior to the experiment and the total score on fifteen reports was taken as a measure of the notebooks to assess laboratory knowledge.
In March the criterion test was administered in two sections. Section I tested chiefly manipulations and was an individual test. Section II consisted of a series of small tests based on the course of study.
About one week later the group pencil and paper test was administered to the three classes in successive class periods. The test consisted of two parts: 1. multiple-choice items, and 2. items matching diagrams with statements.
The following statistical measures were reported for all tests: mean, standard deviation, reliability. For the criterion and pencil paper test the following were also reported: internal consistency of test items with their difficulties. The validities of the items of the pencil and paper test were also reported.
The correlations between the different tests were calculated as a means of appraising the predictive value of each. The simple regression and multiple regression equations and beta coefficients for predicting the criterion from the pencil and paper test were compared. T-scores were tabled for the pencil and paper test as well as derived scores on the basis of a mean of 63 and a standard deviation of 13, designed so as to set 50 as the critical score to cut off 15 percent of the testees.
To compare the ability of the test to predict the upper quarter on the criterion with the lower quarter, a chi-square test of significance was applied.
The following conclusions appear to be defensible:
1. The group pencil and paper test, in predicting the criterion, was significantly superior to other methods,
2. The laboratory notebooks failed significantly to predict the outcomes being tested,
3. The teacher's estimates did not materially assist the pencil and paper test to predict the outcomes being tested,
4. The two tests possess a range of difficulty conforming to the requirements of a good test,
5. The test items having indices of validity of less than .23 contribute little to the predictive value of the pencil and paper test.
6. The pencil and paper test predicts the criterion equally well at either the upper or lower levels. / Education, Faculty of / Graduate
| Identifer | oai:union.ndltd.org:UBC/oai:circle.library.ubc.ca:2429/40877 |
| Date | January 1952 |
| Creators | Chapman, Victor Lennie |
| Publisher | University of British Columbia |
| Source Sets | University of British Columbia |
| Language | English |
| Detected Language | English |
| Type | Text, Thesis/Dissertation |
| Rights | For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use. |
Page generated in 0.0014 seconds