Research Methods and Design Questions and Answers

Answer: B

Purposive sampling is a nonprobability technique. The researcher chooses the units deliberately on the basis of judgement about what the study needs, so not every unit has a known chance of selection.

Simple random, systematic and stratified sampling all give units a known chance of selection, which makes them probability methods. Purposive sampling is common in qualitative studies where particular informants are wanted for what they know.

Answer: A

An increase in alpha, the level of significance, causes an increase in the probability of a Type I error.

Alpha is the probability of rejecting a true null hypothesis, which is the Type I error. Raising alpha from, say, 0.01 to 0.05 widens the rejection region, so the chance of a false rejection grows.

Option B states the reverse. Option C is wrong because alpha directly sets the Type I risk. Option D claims a decrease in Type I error, which cannot follow from a higher alpha.

Answer: C

The assumptions that must be met in the chi-square test are that the data are in frequencies and that all observations are independent, so b and c.

The test compares observed with expected counts of categories, and each observation must be counted once, independently of the others.

Data need not be continuous (a), equal frequencies in all cells are not required (d), and ratio-scale data are not needed (e), since the test works with nominal categories. Options A, B and D contain at least one of these.

Answer: A

Both statements are true, but the reason is not the correct explanation of the assertion.

The hypothesis put to the test is the null hypothesis, so the assertion is true. The alternate hypothesis is accepted only if the null is rejected, which is also true. But the second fact describes the alternate hypothesis, and does not explain why the tested statement is called the null hypothesis.

Option B says it does explain, which it does not. Options C and D call a true statement false.

Answer: C

The correct order of the steps is: state the research hypothesis, state the null hypothesis, select a probability of error level, select and compute the test, and interpret the results, so d, e, c, b, a.

The null hypothesis follows from the research hypothesis, and alpha is fixed before the test is calculated so that the result does not influence the decision.

Options A and B start with interpretation or put alpha before the null hypothesis, and option D puts the null hypothesis ahead of the research hypothesis.

Answer: C

The correct match is x-bar with sample mean, the Greek mu with population mean, rho with probability and sigma with standard deviation of the population, so iv v i ii.

The bar over x marks a sample mean and mu the population mean. Sigma is the standard deviation of the population, and the standard deviation of a sample is written s. The key matches rho with probability, as the paper does.

Options A and B give x-bar as population mean, and option D gives sigma as the sample standard deviation, which is wrong.

Answer: D

Both statements are correct. An experimental design permits inferences about cause and effect, because the researcher manipulates the independent variable and observes the effect it has on the dependent variable while controlling other factors.

The reason is the way in which the design gives that inference. The options that call either statement wrong are rejected. Because the researcher controls the independent variable and assigns subjects, other explanations are reduced, which is the reason cause can be inferred.