While I was looking at Chinese B extended essay samples, I came across a piece of work that had a questionnaire about how students feel about Chinese college entrance exams.The questionnaire asked 40 high school students whether these entrance exams have a negative impact on their happiness, health, and well-being. This made me think about human sciences and the reliability of making generalisations from surveys.

My knowledge question is: To what extent are surveys and questionnaires reliable for making interpretations and generalisations?

Questionnaires, such as the one about entrance exams in the extended essay sample, is a part of the participant-observer methodology in human sciences. Each person who participates in the questionnaire gives a first-person report as an experimental subject. The questionnaire in the extended essay reflects that most students feel that Chinese college entrance exams put a lot of pressure on them, and have a negative impact on their happiness and well-being. The writer of the extended essay, through the use of inductive reasoning, is able to come to the conclusion that these exams are detrimental to most students. However, what are some ways that this methodology can be unreliable? First of all, the writer assumes that these responses are accurate because the experimental subjects offer reasons; however, sometimes first-person testimonies can be undependable because individuals are unwilling to share their true opinions. Or, if the questionnaire is on a sensitive or personal topic, people may not be completely open and honest in their responses. Moreover, since the writer only asked 40 students to do the experiment, this small group of answers could be too limiting for her to draw reasonable conclusions. Another aspect that can make questionnaires unreliable is the intention of the observer. For instance, in the extended essay, the writer is trying to prove that entrance exams have deleterious effects on high school students; thus, she will probably use language that guides the subjects in the direction that she wants. When the observer has a motive, like the writer (she wants to make a claim), he or she may bring subjectivity into the experiment. In this case, the results are likely to be inaccurate, and it would be best if the observer can maintain an analytical distance from the subject matter. Because of this ‘human factor’, interpretations made by human sciences can be biased sometimes.

Results given by human sciences are statistical rather than certainties; and unlike natural sciences, where definitive answers can be given, human sciences usually give us general understandings to explain phenomena in life.  As readers of academic journals/essays/textbooks of interpretations made from human sciences, we should always have a healthy skepticism towards all claims.