Research Exam 2 Practice Problems

Falsification

Two graduate students have made some measurements on a new material. The data points are as shown. To prove their hypothesis the results should lie on the curve shown. The two students excluded the two data points which were off the theoretical curve. What type of misconduct is presented in this case?

-Exempt -Expedited -Full

What are the three classifications for IRBs?

Hawthorne Effect

What is the major threat to the internal validity of an observational research?

(b) A study to evaluate a newly designed wheelchair by asking elderly individuals to use it

Which of the following protocols most likely requires IRB review: (a) A study to determine the effects of a pesticide on a plant (b) A study to evaluate a newly designed wheelchair by asking elderly individuals to use it (c) A study to perform computer simulated stress tests on an artificial hip (d) A study to implant a heart valve in a chimpanzee

d. Personal interview is a method of conducting survey research

Which of the following statement(s) are true about conducting a survey? a. Random sampling procedure is not needed b. Informed consent is not required in survey research c. One should use as many open-ended items as possible d. Personal interview is a method of conducting survey research

c. When applying the results of published research to the clinical practice, it is always better to make sure that your clients have similar traits with those of research subjects (e.g., similar population, age range, etc).

Which of the following statement(s) are true? a. Well-known scientists always conduct well-designed studies, thus peer-reviewed process can be waived when they publish journal articles b. When examining the effects of a high-protein diet program on bone density, it is always better to attribute any beneficial changes during the study to the intervention effects (e.g., reduced BP) c. When applying the results of published research to the clinical practice, it is always better to make sure that your clients have similar traits with those of research subjects (e.g., similar population, age range, etc). d. All of the above

No

Can informed consent be waived since no experiment is conducted in nonexperimental research?

Plagarism

In preparing her thesis introduction, Graduate Assistant Melissa periodically takes multiple sentences (in exactly the same words as were used originally) from some of her sources. (Her attitude is, "I couldn't have written it better myself.") What type of misconduct is presented in this case?

b. Falsification

A PhD student proposed to examine if a novel exercise program would increase more muscle mass than a routine jogging training. He found that three out of the twenty subjects in the experimental group did not increase their muscle mass. He decided to exclude those three subjects' data so the remaining results would support his hypothesis. What type of misconduct is presented in this case? a. Fabrication b. Falsification c. Plagiarism d. Nonpublication of data e. All of the above

Informed consent

A group of undergraduate students planned a research project on the detection of fetal abnormalities in the second trimester, by ultrasound scanning. They collected data from the scan room without informing the mothers. What is this violating?

Misleading authorship

Professor Conan Barbarian is an icon in the field of gerontology. He is the director of the Institute of Gerontological Research at Jellystone University, a highly funded research lab with lots of graduate students. Professor Barbarian requires that he be listed as an author on all manuscripts based on research completed in his lab. What is this an example of?

e. Both (b) and (c)

To control the quality of research, what are the most important internal and external control mechanisms? a. Self-disciplined laboratory guidelines b. Peer-reviewed publication process c. IRB (Institutional Review Boards) d. Both (a) and (b) e. Both (b) and (c)

Accept Ho; Independent t-test; 3-way ANOVA

A PT has developed a new exercise program that she thinks can help older people to reduce their body fat. She wanted to determine if subjects who received the new intervention would reduce more body fat over one year after the training compared to those receiving conventional PT. Eighty subjects were randomly assigned to each of the two programs (new vs. conventional). Q1. Level of the significance was set at 0.05. She got a p value that is 0.06. Her decision on taking hypothesis? Q2. What is the most appropriate statistical analysis to compare the effectiveness of the two programs? Q3. Later she learned that the diet and gender may also affect the changes in body fat and interact with the intervention effects. What is the statistical analysis to test for the effects of diet, gender, the type of interventions and their interaction effects on the changes of body fat?

c. does not manipulate variables and hence cannot establish cause and effect

A basic characteristic of correlational research is that it: a. can deal with only two variables b. cannot predict future performance c. does not manipulate variables and hence cannot establish cause and effect d. does not need both measures on all participants for two variables to be correlated

(c) Exempt as the study presents no greater than minimal risk and does not collect identifying information.

A large HMO has teamed up with several other large HMOs to study falls and footwear. The investigator proposes to review 5,000 medical records across 10 institutions of individuals injured during the last three years from a fall, so he has enough power to detect whether there is a greater number of falls among people who wear clogs. Because individuals wear so many different types of shoes, he will need to review a large number of records to find those few individuals who fit this category. From the medical records, he plans to extract the type of shoe the individuals wear. Which of the following is likely to be the level of review? (a) Expedited review required, as the study presents no greater than minimal risk but involves the use of medical records that contain sensitive information (b) Full IRB review because the study exposes subjects to greater than minimal risk because the use of private information from multiple institutions. (c) Exempt as the study presents no greater than minimal risk and does not collect identifying information.

Static Group Comparison

A researcher designs a study to determine if an intervention will change participants' attitudes toward eating Hershey bars for breakfast. Assume the researcher has a previously validated instrument to measure the dependent variable (i.e., attitude toward eating Hershey bars for breakfast). The researcher takes two intact groups of adults: one working at a telemarketing firm one working at a university library. The group at the telemarketing firm receives a three-hour class where the goal is to change attitudes, and then both groups take the attitude inventory. When the results are analyzed, there is no difference between the two groups. What is the experimental design?

ANCOVA; Two-Way ANOVA

A researcher plans to compare the effects of two methods of relaxation training on free-throw shooting. Unfortunately, one group of his subjects has higher anxiety level compared to another group. Thus, the researcher feels that subjects' baseline anxiety level may affect their performance of free-throw shooting in addition to the relaxation training. Q1. What is the most adequate statistical analysis to use for testing the difference in free-throw shooting due to the two methods of relaxation? Q2. Later, the researcher decides to examine the effects of anxiety level on free-throw shooting in addition to relaxation training. So he stratified the anxiety level to high, med, and low. He wants to investigate how the free-throw shooting is affected by different relaxation methods, anxiety levels and the interaction between the two. What is the most adequate statistical analysis to use?

b. Increase the effect size by reducing the variance of the data (e.g., reducing any possible measurement error)

A researcher wanted to increase the power of her study. She could: a. Reduce the sample size b. Increase the effect size by reducing the variance of the data (e.g., reducing any possible measurement error) c. Set the alpha level at 0.01 instead of 0.05 d. Both (a) and (b) e. (a), (b) and (c)

(d) developmental

A researcher wants to investigate how bone density changes from childhood, adolescence to adulthood. What type of research he is conducting? (a) epidemiological (b) observational (c) experimental (d) developmental (e) correlational (f) case studies

(a) Correlation; (b) Partial correlation

An investigator has developed a new algorithm that can estimate the Oxygen consumption (VO2) of an individual based on his/her heart rate data. To validate this algorithm, this researcher used a metabolic cart to simultaneously record subjects' VO2 and heart rate. Q1. What is the most appropriate statistical analysis to quantify the validity of this new algorithm in estimating the VO2 using the heart rate data? (a) Correlation (b) Partial correlation (c) Dependent t-test (d) Two-way ANOVA (e) Post-hoc (f) ANCOVA Q2. If there are other factors (e.g., anxiety level, fitness level) may affect the relationship between the heart rate and VO2, what is the most appropriate statistical analysis to determine the relationship between the two? (a) Correlation (b) Partial correlation (c) Dependent t-test (d) Two-way ANOVA (e) Post-hoc (f) ANCOVA

Misleading authorship

Four friends decide to work together on a research project during the vacation. One of them went abroad during the vacation and did not contribute to the research. The friends include all 4 names in a presentation made at a scientific conference. What type of misconduct is presented in this case?

(c) Pre-experimental - static group comparison design

Jane wants to examine how three different learning styles affect students' scores on the final exam. She planed to recruit three groups of subjects. However, it is difficult to do group randomization because of the conflicts in students' schedules. She then used students from three different sessions of the mathematics course instead. Since there would be substantial learning effect from the testing, she only issued a test after the intervention and compared the final exam grades among the three class sessions. Q. What type of experimental design is used? (a) True experimental - posttest only design (b) Quasi-experimental - pre and post nonrandom design (c) Pre-experimental - static group comparison design (d) Pre-experimental - one shot design

One-Way ANOVA; ANCOVA

OBJECTIVE: To determine the amplitude of the electromyographic activity of trunk muscles during Pilates exercises in women with chronic low back pain (LBP) (n=20), women who have recovered from the chronic low back pain (n=20), and women who never experience low back pain (n=20). Q. Most appropriate statistical analysis to test for the differences in muscle activation among the three groups? Q. If we suspect the thickness of the skin and fat would affect the electromyographic data other than the presence of low back pain, then what is the most appropriate statistical analysis to test for the differences in muscle activation among the three groups?

d. Placebo effect

Randomization is a good way to control or reduce the effects of confounding factors or threats that reduce a study's internal validity. However, randomization CANNOT reduce or minimize the effects from: a. Maturation b. History c. Selection bias d. Placebo effect

Plagarism

Students are required to prepare a research proposal during their undergraduate program. Peter developed the idea for his project and discussed with a friend, John. Several months later, he found that his idea had been submitted as a research proposal by John without his knowledge. What type of misconduct is presented in this case?

f. (b) and (c)

Susan set the level of significance (α) at 0.10 instead of 0.05 in his study. Which of the following statements is correct? a. The power of study reduces b. Type II error reduces c. Type I error increases d. Getting harder to detect a significant difference e. (a) and (c) f. (b) and (c)

d. instrument accuracy

Test administrators usually become more skilled with practice in using the laboratory equipment, which could cause differences between data obtained in the pre-test and post-test. This threat to internal validity is an issue of: a. statistical regression b. testing c. history d. instrument accuracy e. selection bias

d. random assignment to groups

The one characteristics of true experimental designs that pre-experimental and quasi experimental designs do not have is : a. the use of a control group b. a pretest and posttest c. the application of parametric statistics d. random assignment to groups