Research Methods and Design Questions and Answers

Answer: B

Scientific theories arise through inductive reasoning.

A theory is a broad explanation built by generalising from many specific observations and tested results, which is inductive reasoning: from particular cases to a general principle. Deduction is then used in the opposite direction, to derive specific predictions from the theory that can be tested.

So theories are tested deductively but arise inductively. A theory is a wide generalisation, not a 'least' one. And a scientific theory must be open to disproof by contrary evidence; one that could never be disproved would not be scientific.

Answer: B

The variations cannot ideally be reduced to zero, however many times the experiment is repeated.

Biological systems and measurements always carry random variation: individuals differ, conditions fluctuate slightly, and instruments have limited precision. Repeated readings that differ but agree statistically show this normal random variation, not a mistake.

Such variation does not by itself indicate an error in observation or calculation, nor a failed or badly designed experiment. Repeating the experiment until some readings coincide would not remove it; more repetitions only estimate the mean and spread more precisely.

Answer: D

The appropriate scale is the nominal scale.

Suburbs, city, town, metro and villages are names of categories. They can be counted and compared for equality, but the list has no natural order, no meaningful differences and no true zero, which is the definition of nominal data.

An ordinal scale needs a ranking (such as 'low, medium, high'); an interval scale needs equal, meaningful differences (such as temperature in degrees Celsius); a ratio scale needs a true zero as well (such as income). None of these fit place-of-residence labels.

Answer: D

In a probability sample every unit has a known chance of selection. Stratified sampling divides the population into strata and draws random samples from each, so it is a probability method. Convenience, quota and judgement sampling choose units by the researcher's ease or choice, so they are non-probability methods. The answer is stratified sampling.

Answer: A

A Type I error is made when a true null hypothesis is rejected. It is the error of finding an effect that is not there, and its probability is the level of significance, alpha (for example 0.05 or 0.01).

Accepting a true null hypothesis is a correct decision. Rejecting a false null hypothesis is also a correct decision, whose probability is the power of the test. Accepting a false null hypothesis is the Type II error, whose probability is beta.

Answer: A

Statements (i), (ii) and (iv) are correct. The variance is the average squared deviation and is proportional to the average squared difference between all pairs of observations. It is smallest when the deviations are measured from the mean, because the sum of squared deviations is least about the mean. It is also a measure of dispersion of a distribution.

Statement (iii) is wrong: the Greek letter sigma denotes the standard deviation, while the variance is written as sigma squared. So 'all are correct' and the options with (iii) are wrong.

Answer: A

Both (A) and (R) are true, and (R) is the correct explanation. A correlation measures how two variables vary together, so it can be shown only when there is variability in both.

If one variable is a constant it has zero variance. The correlation coefficient has the standard deviations of both variables in its denominator, so it becomes a division by zero and is undefined. That is exactly why variability is needed, so (R) explains (A). The options that deny either statement, or that call (A) true and (R) false, are wrong.

Answer: C

The correct matching is (a)-(iii), (b)-(i), (c)-(iv), (d)-(ii). Planned methodical watching of a subject or situation is observation. Noting the opinions of subjects in a group is focus group discussion. A document containing a set of questions is a questionnaire, and a set of structured questions whose answers the researcher records is an interview schedule used in an interview.

The other codes put the questionnaire against the interview or observation against the focus group, which reverses the definitions. So option C is the answer.

Answer: B

Saying that nine year olds are taller than seven year olds compares two different age groups at the same time, which is the way of a cross-sectional study. A longitudinal study would follow the same children over the years, and a time series study measures one thing at successive times. An experimental study would involve a treatment and a control group, which is not the case here. The answer is a cross-sectional study.

Answer: A

Ex post facto means 'from what is done afterwards'. In ex post facto research the researcher studies an event or its effects after it has taken place, and tries to trace the causes, without being able to manipulate the variables. So the research is carried out after the incident. The other options describe research done before or during the event, or in anticipation of it. The answer is that the research is carried out after the incident.