Critique of a research article titled “Does bulling cause emotional problems? A prospective study of young teenagers” authored by Lyndal Bond, John B Carlin, Lyndal Thomas, Kerryn Rubin and George Patton, and published in the British Medical Journal 2001, volume 323. The authors set out to investigate if there was a causal relationship between being bullied and the resulting self reported emotional problems in the secondary school teenagers being followed up for 2 years. At the end of the study they reported that being bullied and having poor social relations significantly resulted in anxiety or depressive symptoms in the school children.
A critical look at how the title related ...view middle of the document...
Duration of follow up
The authors stated that the cohorts were followed up for a period of two years. The temporal association between the probable time of victimization and the resultant emotional problems appears too distanced. The time between the first wave of study in year 8 and the last wave of the study in year 9 has a separation of more than 20 months. A single episode of victimization leading to depressive symptoms would, theoretically, have resolved within this time frame. I would suggest the duration of the study to last for one year in which the three waves would be at the beginning of year 8, six months after and the emotional problems recorded at one year.
Rate of follow up
The follow up rate of the students this study appears adequate. However, the individual assessments are considered too far apart as I stated in the preceding section.
The sample of 2860 (79%) out of 3623 students had at least one wave of data while 2559 (71%) provided the data for analysis. This means that the attrition rate was as high as 29% of the sampled population. This should be of concern, especially when it appears that those who dropped out from the study were significantly biased towards having attributes of students being bullied. However, it must be admitted that the authors were meticulous enough to account for the drop-outs. The authors gave the impression that all randomly sampled individuals who were present in the school participated in the study. They did not indicate the number of students who declined to participate, their reasons for doing so and how their replacements were carried out.
Measurement of exposure data
The measurement of bullying in two waves of the study and the survey of absent students at a later date or by telephone was a good practice. The authors stated that answering yes to the four questions of victimization is taken that the student was victimized. However, they left a little ambiguity in not stating if classification as a case victimization was extended to a student on the basis of answered yes to just any one of the four questions or all the four of the questions. In other words, the...