The assessment of programmers and analysts. [in Proc. 5th National Conf. of the Computer Society of Canada, 1966, 163-166.]
|
 |
|
Post a Comment
|
|
|
|
|
CONTRIBUTORS:
|
|
|
JOURNAL:
|
Computing Reviews,
10(7),
303 -
303.
|
|
|
|
YEAR:
|
1969
|
|
PUB TYPE:
|
Book Review
|
|
SUBJECT(S):
|
programmer assessment; testing; BPKT
|
|
DISCIPLINE:
|
Computer Science
|
|
HTTP:
|
|
|
LANGUAGE:
|
English
|
|
PUB ID:
|
103-427-468
(Last edited on
2006/06/07 15:40:02 GMT-6)
|
|
ABSTRACT:
This is a report on the investigation of programmer assessment through the use of several types of tests at the University of Southern California.
A long-range study is being conducted by the university for the US Office of Naval Research and the Naval Command Systems Support Activity. It consists largely of job analysis to determine the mental and manipulative behaviors required for computer programmers and systems analysts, aptitude test development for selecting novices for entry positions, and proficiency test development to evaluate the levels of proficiency of experienced individuals in this field.
The report discloses that the job analysis phase produced 17 distinct factors or job dimensions which provide a set of concepts from which the testing phases will evolve. The paper, however, focuses upon proficiency tests only. Of the two tests in this battery, the Basic Programming Knowledge Test (BPKT) has been developed and validated. It consists of two sections: one comprising 75 multiple-choice questions of a generalized nature, so that they are relatively machine- and language-independent. The other consists of 25 multiple-choice questions built around a hypothetical computer system, including a hypothetical language, so that coding and debugging of programs can be measured. While .90 is given as the reliability coefficient for internal test consistency, the report does not provide additional validation data. Apparently, validation is a continuing activity.
This is essentially an abstract rather than a full-scale
technlcal report. Consequently, psychometrists, behavioral
scientists, personnei administrators, and others interested
in measuring mental activities of programmers and analysts will only become aware of the project research. Awareness sometimes causes frustration, and rarely does it result in the acquisition of test instruments and the replication of the experiment.
However, the project is a worthy one and hopefully the tests will be fully validated and made available so those of us not in a military establishment will profit. Regrettably, academic wheels, while they do not necessarily move in circles, are known to move with remarkable pokiness. It should not be necessary to call a protest meeting to get the project completed before fourth-generation systems make the test batteries relatively useless in the presence of higher-order user languages.
L. C. Silvern, Los Angeles, Calif.
|
|
|
|
STATISTICS
|
|
Click on # to view
|
|
Citations
|
|
0
|
|
References
|
|
0
|
|
Comments
|
|
0
|
|
Quality
|
|
0/0.00
|
|
Interest
|
|
0/0.00
|
|
View(er)s
|
|
2/149
|
|
|
|
|
|
|
| Prev |
Next |
|