Quantitative design: Ungraded review questions

Transcript:Quantitative design: Ungraded review questions:
Quantitative design: Ungraded review questions Which of the following is NOT true about expost facto design? It is not experimental and data are collected after the fact. it is causal-comparative and it is considered the best way to make causal inference It is used when randomization is not possible. Which of the following is NOT true about observational research? It is non-experimental Correlations between variables are studied The researcher must observe the subjects in the field. One of the weaknesses is that correlation does not necessarily imply causation. Traditionally, most Evangelicals and Republicans work hand in hand. Professor Trump would like to conduct a study to affirm that religious conservatism (independent variable) tend to push people to political conservatism (dependent variable). His research assistant Sarah said that this so-called causal study is problematic. What is/are the flaw(s)? Even if two variables are statistically correlated, it may not be a causal relationship. There are many other factors influencing political orientation, but Trump’s study does not take them into account. It is possible that the causal link is reversed (political position  religious belief) or bidirectional (political position  religious belief). All of the above After examining international assessment scores (PISA, TIMSS) and the GNP per capita of many countries, Dr. April May found that there is no relationship between test performance and economic achievement. She concluded that better test skills do not necessarily lead to better job performance. What is her mistake (Choose the best answer)? Ecological fallacy: She infers from country-level data (summary data) to individuals. International assessment data are unreliable. Those tests are culturally biased. PISA and TIMSS do not use multi-stage sampling. All of the above Dr. Noel Bobbie wants to conduct a nationwide survey. Which of the following sampling scheme should she use? Convenience sampling Simple random sampling Multi-stage sampling All are good. Dr. Kent Clark, who works for the state government, wants to ensure that the sample collected from California can represent the state population. Assume that he has access to the state census data. What should he do? Check the population to make sure that it is normally distributed. Check whether simple random sampling is properly implemented. Check the sample properties against the population properties. All of the above Which of the following is NOT a reason for doing a pilot study? Spot errors and ambiguity in wording. Spot errors in the interface design and