In controlled comparison research, the observed relationship between an independ

Question

In controlled comparison research, the observed relationship between an independent variable (X) and a dependent variable (Y) may be caused by some other, uncontrolled compositional difference (Z). So we always ask, “How else, besides the independent variable, are the groups I am comparing not the same?” For each of the three hypotheses below: (i) Think up a plausible alternative causal variable. (ii) Describe how this variable might affect the relationship between the independent variable and the dependent variable.
Example. In a comparison of countries, those with decentralized governments have higher percentages of citizens who trust the government than do countries with centralized governments.

(i)   Plausible alternative variable: size of country’s population

(ii)  Countries with decentralized governments might also have smaller populations than countries with centralized governments. If smaller countries also have higher percentages of trusting citizens than do larger countries, then differences in size, not differences in decentralization-centralization, could explain the relationship between the independent and dependent variables.

A.   In a comparison of individuals, married people are more likely than unmarried people to be Republicans.

B.   In a comparison of individuals, those who are citizens of countries with government-funded health care have longer life expectancies than do citizens of countries with privately funded health care.

C.   In a comparison of individuals, southerners are more likely than non-southerners to approve of spanking as a way to discipline children.

Explanation / Answer

A. i) Plausible alternative variable: total income of the family.

ii) If unmarried people have more income leading to more stability, then family income would have explained the relationship between voting preference and marital status.

B. i) Plausible alternative variable: cost of treatment

ii) If privately fundedhealth care would have been successful in providing good healthcare in low cost, then, this variable would have explained the relationship between life expectancies and type of healthcare.

C.i) Plausible alternative variable: Number of elementary schools

ii) If southerners have equal or more number of elementary schools to put their kids at the first stage, then this varible alone can explain the relationship between region and approval of spankin to discipline children.

Related Questions

Navigate

• Browse (All)
• Browse I
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.