Study Design Practice Questions

Q: A researcher is examining the relationship between socioeconomic status and IQ scores. The IQ scores of young American adults have historically been reported to be distributed normally with a mean of 100 and a standard deviation of 15. Initially, the researcher obtains a random sampling of 300 high school students from public schools nationwide and conducts IQ tests on all participants. Recently, the researcher received additional funding to enable an increase in sample size to 2,000 participants. Assuming that all other study conditions are held constant, which of the following is most likely to occur as a result of this additional funding?

Decrease in standard error of the mean. ***Decrease in standard error of the mean*** - **Increasing the sample size** (n) leads to a **decrease in the standard error of the mean** (SEM), which is calculated as σ/√n. - A smaller SEM indicates that our sample mean is a more **precise estimate** of the true population mean. *Increase in risk of systematic error* - **Systematic error** is related to flaws in study design or implementation and is not directly affected by an increase in sample size. - A larger sample size generally helps in detecting a true effect if one exists, but does not inherently introduce or correct systematic bias. *Increase in range of the confidence interval* - An **increase in sample size** typically leads to a **narrower confidence interval**, not a wider one, because the standard error of the mean decreases. - A narrower confidence interval implies greater precision in estimating the population parameter. *Decrease in standard deviation* - The **standard deviation** is a measure of the data's spread within a sample or population and is an intrinsic characteristic of the data itself. - Increasing the sample size typically does not change the true standard deviation of the population; it only provides a **more accurate estimate** of it. *Increase in probability of type II error* - An **increase in sample size** generally leads to an **increase in statistical power**, which in turn **decreases the probability of a Type II error** (failing to reject a false null hypothesis). - A larger sample makes it easier to detect a true difference or effect if one exists.

Q: A case-control study is conducted to investigate the association between the use of phenytoin during pregnancy in women with epilepsy and the risk for congenital malformations. The odds ratio of congenital malformations in newborns born to women who were undergoing treatment with phenytoin is 1.74 (P = 0.02) compared to newborns of women who were not treated with phenytoin. Which of the following 95% confidence intervals is most likely reported for this association?

1.34 to 2.36. ***1.34 to 2.36*** - A **95% confidence interval** that does not include **1.0** indicates a statistically significant association, consistent with the given **p-value of 0.02** (which is less than 0.05). - This interval contains the **point estimate (odds ratio of 1.74)**, making it the most plausible range for the true effect. - The confidence interval must always encompass the point estimate from which it is derived. *0.36 to 0.94* - This confidence interval is **entirely below 1.0**, suggesting a protective effect, which contradicts the given odds ratio of **1.74** indicating an increased risk. - This interval does not contain the point estimate of **1.74**. *1.75 to 2.48* - While this interval indicates an increased risk and does not include 1.0, it **does not contain the stated odds ratio of 1.74**, as its lower bound is 1.75. - A confidence interval must always encompass the point estimate from which it is derived. *0.56 to 1.88* - This confidence interval **includes 1.0**, which would imply no statistically significant association between phenytoin use and congenital malformations. - This contradicts the given **p-value of 0.02**, which indicates statistical significance. *0.83 to 2.19* - This confidence interval **includes 1.0**, suggesting **no statistically significant association**. - This contradicts the given **p-value of 0.02**, which demonstrates statistical significance.

Question 1

A researcher is examining the relationship between socioeconomic status and IQ scores. The IQ scores of young American adults have historically been reported to be distributed normally with a mean of 100 and a standard deviation of 15. Initially, the researcher obtains a random sampling of 300 high school students from public schools nationwide and conducts IQ tests on all participants. Recently, the researcher received additional funding to enable an increase in sample size to 2,000 participants. Assuming that all other study conditions are held constant, which of the following is most likely to occur as a result of this additional funding?

Accepted Answer

Decrease in standard error of the mean

Answer

Increase in risk of systematic error

Answer

Increase in range of the confidence interval

Answer

Decrease in standard deviation

Answer

Increase in probability of type II error

Question 2

A case-control study looking to study the relationship between infection with the bacterium Chlamydia trachomatis and having multiple sexual partners was conducted in the United States. A total of 100 women with newly diagnosed chlamydial infection visiting an outpatient clinic for sexually transmitted diseases (STDs) were compared with 100 women from the same clinic who were found to be free of chlamydia and other STDs. The women diagnosed with this infection were informed that the potential serious consequences of the disease could be prevented only by locating and treating their sexual partners. Both groups of women were queried about the number of sexual partners they had had during the preceding 3 months. The group of women with chlamydia reported an average of 4 times as many sexual partners compared with the group of women without chlamydia; the researchers, therefore, concluded that women with chlamydia visiting the clinic had significantly more sexual partners compared with women who visited the same clinic but were not diagnosed with chlamydia. What type of systematic error could have influenced the results of this study?

Accepted Answer

Reporting bias

Answer

Detection bias

Answer

Lost-to-follow-up bias

Answer

Ascertainment bias

Answer

Response bias

Question 3

A case-control study is conducted to investigate the association between the use of phenytoin during pregnancy in women with epilepsy and the risk for congenital malformations. The odds ratio of congenital malformations in newborns born to women who were undergoing treatment with phenytoin is 1.74 (P = 0.02) compared to newborns of women who were not treated with phenytoin. Which of the following 95% confidence intervals is most likely reported for this association?

Accepted Answer

1.34 to 2.36

Answer

0.36 to 0.94

Answer

1.75 to 2.48

Answer

0.56 to 1.88

Answer

0.83 to 2.19

Study Design — MCQs

Study Design — MCQs

On this page

Practice by Chapter

Want unlimited practice?