Tuesday, March 4, 2008

How to Be a Statistics Sleuth: Seven Critical Components (Utts 2005)

Component 1: The source of the research and of the funding

Component 2: The researchers who had contact with the participants

Component 3: The individuals or objects studied and how they were selected

Component 4: The exact nature of the measurements made or questions asked

Component 5: The setting in which the measurements were taken

Component 6: Differences in the groups being compared, in addition to the factor of interest

Component 7: The extent or size of any claimed effects or differences

---
Now, let's analyze my project with these seven components!

Component 1: NUS DBS Plant Systematics Laboratory, with honest motivations, seeking honest results.

Component 2: Assoc Prof Hugh T. W. Tan appealed to students to do the survey during his SSS1207 – Natural Heritage of Singapore class on the 21st of February 2008. A possible biased would be the influence the researcher might have on the participants, as participants are more likely to answer according to the desires of the researcher. Therefore having a conservation biologist conduct the survey would perhaps have some effect on the students.

Component 3: It is possible that students taking this module, do so, because they like nature, or have an interest to learn more about it. Hence, this sample selected is biased, and it would not be representative of the undergraduate student population in NUS.

Component 4: Some students avoided revealing their family income and religion and left blanks here and there. The questions asked were related to behaviour, attitude, knowledge and demographics.

Component 5: Survey was conducted before break time. Students might have been tired and rushing to finish the survey in order to have time to visit the toilet or get some food. This might have affected the quality of answers. But students overall appeared very cooperative.

Component 6 is not applicable.

Component 7 would be discussed in time to come.

Reference:
Utts, J. M. 2005. Seeing Through Statistics. Third Edition. Brooks/Cole

No comments: