..................................................................................................................................................................................................................
..................................................................................................................................................................................................................
Power quality affects teacher well being and student behaviours in three Minnesota Schools
Comments on a poor study done by leading Canadian alarmist Dr. Magda Havas
This is another "study" conducted by Magda Havas. This "study" is similar to one that Havas did at a Toronto school about the effects of "dirty electricity". In both "studies" teachers filled in a daily questionnaire about their EHS like symptoms including: "Headaches, general weakness, dry eyes/mouth, facial flushing, asthma, skin irritations, overall mood including depression and anxiety", etc., and another questionnaire about the behavior of their students. The purpose of the "study" was to analyze the effectiveness of so called Graham Stetzer filters in improving their symptoms and student behavior. The filters supposedly work by significantly reducing the level of "dirty electricity". Dummy filters were installed for 2 weeks, "real" filters for 4 weeks, and dummy filters for the last 2 weeks.
The "study" was flawed from the outset because it was "single blinded" (Havas knew when real and dummy filters were installed), and the number of "dummy" filters was not the same as the number of "real" filters (285 "dummy" vs. 541 "real"). This means that some teachers and students might easily have become aware of the difference between the "real" and "dummy" phases of the trial. With such a flawed methodology both teachers and students might get an unconscious clue to the experimenter's desired response. In the study's abstract, Havas summarizes her conclusions: "Of the 44 teachers who participated 64% were better, 30% were worse, and 6% did not change. Behavior of high school students did not improve but elementary/middle school students were more active in class; more responsive, more focused; had fewer health complaints; and had a better overall learning experience."
In addition to the fact that this study was not double blind, there is a serious exaggeration is the interpretation of the charts and the statistical methods she uses to reach her conclusions. How does Havas conclude that 64% of teachers were better? See the above chart for Elementary, Middle, and High School Teachers from the paper. The left chart shows overall response from all teachers on each of 44 "symptoms". The right chart shows each 44 teacher's overall response to the 44 "symptoms". Note that the dominant color on both charts is yellow for "same".
But here Havas performs an unusual massaging of the numbers. On the left chart, on the first line for "headache", around 22% of teachers reported feeling better, around 70% felt the same, 8 percent felt worse. The fact that slightly more teachers felt "better" means that this symptom counts as a "hit". She does this for all 44 symptoms. According to this analysis, 79% of the symptoms were improved. The same methodology applies to the right chart. Teacher 1 felt better on about 73% of the "symptoms", and the same on 27%. Teacher 18 felt better on around 10% of symptoms, the same on 85% of symptoms, and better on 5%. She also counts as a hit. This flawed analysis leads to her misleading conclusion that ""Of the 44 teachers who participated 64% were better, 30% were worse, and 6% did not change".
This "methodology" is just amplifying a weak result. In the "teachers" charts, it is clear that the predominant color is yellow or "same" and represents about 80% of all responses. Of the remaining 20 % of teacher's responses there is a slight preponderance of "better". Also revealing is the chart for high school students. According to her methodology, on the left chart 69% were worse, 31% better, and on the right chart 53% were worse, 35% better. So by her own flawed methodology, her overall conclusion should have been that while the teachers were improved, the high school students were "worse". But in her abstract she simply says the students "did not improve". So in addition to not using a double blind methodology, and using a flawed statistical methodology of amplifying a weak result, she is also highlighting only the portion of her "results" that agree with her foregone conclusion. Overall this deeply flawed "study" proves nothing.
Except where noted all images on this web site are taken from the Wikipedia commons
All trademarks are the property of their respective owners Copyright 2009 EMF & Health