Critique of academic article analysis

Critique of Article: Health and Safety in the Engineering Classroom

1 The type of research is descriptive, to elicit statistics on how prevalent safety and health instruction is in the normal college engineering curriculum. 2. Yes, the problem is clearly stated. "The Professional Engineer's Code of Ethics includes the responsibility 'to hold paramount the public safety, health and welfare,' and yet several recent reports suggest that few undergraduate engineering programs include any structured course material relevant to identifying environmental threats to health and controlling occupational and public health and safety hazards." (Introduction, p. 1, par. 2, lines 1-4) 3. Yes, the research questions are defined, albeit indirectly through the use of tables and "subsets" of the questions asked. One table lists the reasons why professors do include instruction on health and safety in their classes, while the other addresses the reasons they would not include such instruction. While the questions themselves are not spelled out, the information elicited is clearly ranked on these tables (Tables 1 and 2, pages 2-3, pars. 1-4). 4. No, the problem is not deduced from theory but from results of other studies.(Introduction, p. 1, par. 2, line 3, and p. 6 list). 5. N/A. This is not an experimental or relational design, so no independent variables were defined. 6. One dependent variable, as a source of possible error, was that one-third of the respondents previously had attended a National Institute for Occupational Safety and Health-sponsored workshop for engineering faculty (Discussion, p. 4, par. 1, lines 2-3).These faculty, because of their attendance, may have had a stronger focus on safety issues than had all respondents either attended these workshops or been drawn from non-attendees. 7. Yes, the surveys sent to the respondents covered both negative and positive situations, and gave respondents a range of response from "mildly" to "strongly" positive, to neutral or negative. 8. Yes, the population studied was clearly specified. "A questionnaire probing faculty interest in and instructional commitment to occupational and public health and safety was mailed to 324 undergraduate engineering professors in 112 U.S. colleges of engineering." (Methods, p. 1-2, par. 4) 175 surveys were returned, of which 18 were excluded because the professors had either retired or left the university setting. Of the 157 that remained, 97% of faculty responding were from undergraduate programs accredited by the Accreditation Board for Engineering and Technology, representing 65 colleges of engineering (Methods, p. 2, par. 1, lines 13-16). 9. Yes, the sampling methods were clearly outlined. The samples were divided into three groups; one group had attended National Institute for Occupational Safety and Health workshops; one group was chosen from the same universities and fields as those previous, only from non-attendees; and the third group was random (Methods, page 2, par. 1, lines 1-5). 10. This was not experimental research, so except for subgroups within the large sample (Methods, p. 2, par. 1, lines 14-16), there were no control or comparison groups. 11. N/A. This was not experimental research, so there were no treatments. 12. No, there was no replication of this study within the study. 13. No, there was no alpha level specified a priori. 14. Yes and no. The method was described in some detail, but the actual questionnaire was not included in the report. While a questionnaire could be formulated from the answer subsets given in Tables 1 and 2, such a questionnaire would not necessarily be identical to the one used in this study. A researcher would have to procure the exact survey in order to replicate this study. 15. Yes, the sample size was described. In the three subgroups surveyed, fifty-six undergraduate professors were from Group 1, the group that had attended National Institute for Occupational Safety and Health seminars; 51 from Group 2, the group from the same universities and fields but who had not attended the seminars; and 50 from Group 3, professors chosen at random from faculty lists both in the 1992 Directory of Engineering and Engineering Technology Undergraduate Programs and the 1992 Love Joy's College Guide, with faculty names being selected from catalogs on microfiche (Methods, p. 2, par. 1, lines 2-10). 16. N/A. There were no control groups in this study, and the same survey was given to all three groups mentioned. 17. No. Except for the "Discussion" about whether the professors who'd attended the National Institute seminars, there was no mathematical indication of reliability factors. 18. No, no evidence of the validity of the measurement was given, and since the precise survey questions are never mentioned, there is no material given upon which to base a measurement of bias, semantics, or other mitigating factors. Since the sample is relatively small, its validity can also be called into question. 19. Yes, the conclusions mentioned on pp. 4-5 (Discussion) were consistent with the obtained results; the overwhelming motivation of faculty who include safety and health instruction in their classes appears to come primarily from within-their own beliefs and ethics-rather than from any external sources such as governing/accreditation bodies. 20. Yes, the generalizations seemed to be confined to the population from which the author drew the sample. However, it should be stated that including data from student respondents to a Tufts University study in results, rather than in background and/or survey information, seems problematic and may unnecessarily "skew" the recommendations and interpretation of data (Discussion, p. 5, par. 1, lines 3-8). 21. In my opinion, this is not a significant study. It surveys such a small number of both engineering schools and especially such a small number of faculty that, statistically speaking, I believe this study would have to be repeated several times to get a true overview of where college engineering faculty "stand" on the matter of safety education in their curricula.