Exam 14: a Statistical Tools for Answering Research Questions

Fix It Dr. Sikorski wants to know if technology is dumbing people down. She notices that years ago, before cell phones were widely used, people had to memorize important phone numbers. In thinking about this, she realizes that she remembers her best friend's phone number from high school, but cannot remember her best friend's phone number today. She hypothesizes that our reliance upon technology to store information for us has produced "digital amnesia." To test whether technology impairs memory she the first recruits 30 undergraduates who come to the lab. The participants are given a list of 20 items they must purchase at a grocery store. The names of the items are spoken by the researcher, and the participants' progress through the three treatment conditions-digital, traditional, and control-in the same order. In the digital condition they are told they may make a list of the items using their phone. In the traditional condition they are told they may make a list using a pencil and paper. In the control condition they are not permitted to record the list. After the participants were read the 20 items they were given a distractor task to prevent rehearsal. One minute later they were asked to speak the items they were to purchase. Participants were not permitted to use their notes to recall the items. The study uses a within-subject design. To ensure consistency in the procedure, the same 20 items were used in each condition. t-tests were performed to examine group differences in the mean number of items recalled. The results support Dr. Sikorski's hypothesis. The participants recalled significantly fewer items during the digital condition than in the traditional condition (t(29) = 7.23, p < .05) and the control condition (t(29) = 5.42, p < .05). She also found that fewer items were recalled in the traditional condition than in the control condition (t(29) = 4.28, p < .05). Dr. Sikorski concludes that memory aids, either digital or traditional, are ultimately detrimental to memory. What do you think of how Dr. Sikorski came up with her research question? What issues do you see with design of this study? Fix them!

Fix It A recent study indicated that people are becoming more supportive of recreational marijuana use. But Dr. Coutley believes that the marijuana produced today is more harmful than marijuana produced in the 1970s. He therefore decides to conduct a study to test his hypothesis. Since Dr. Coutley works for the DEA, he is able to obtain 10 samples of marijuana seized in 1972 and 10 samples of marijuana seized in 2013. To assess potency, Dr. Coutley analyzes the THC content of the marijuana samples. He also assesses the behavioral impact of each sample. Sixty individuals, half of whom are regular marijuana users, were randomly assigned to receive the 1972 marijuana or 2013 marijuana. The participants were told whether they were receiving the 1972 or 2013 drug prior to its administration so each person could make a final decision as to whether to continue their participation in the study. Only two of the 30 marijuana users withdrew from the study. Seven of the 30 nonmarijuana users withdrew from the study. All of the participants who withdrew were assigned to the 1972 drug condition. This resulted in a final sample of 51 participants. Fifteen minutes after receiving the drug the participants completed The Drug Effects Questionnaire, a self-report measure that assesses how "high" one feels. A t-test was used to compare THC content in the two samples. The results revealed that the 2013 marijuana was significantly more potent than the 1972 marijuana (t(9) = 5.32, p < .05). The behavioral effects of the two samples also indicated that the 2013 marijuana was more potent (t(50) = 4.82, p <.05). Dr. Coutley concludes that a linear relationship exists between the year of marijuana production and potency of marijuana. He uses this study to advocate for the nonlegalization of marijuana in his state. Aside from potency, what other factors/biases may have influenced how high participants reported feeling? Do you agree with the analyses that Dr. Coutley performed? Fix it!

Fix It One day while watching television Dr. Ortiz noticed that much of the programming was negative. The local news ran numerous stories about crime and another network channel was playing a marathon of a popular crime drama. Based on this observation, Dr. Ortiz hypothesized that people who watch more television may experience higher levels of perceived threat. To that end, he administered a survey to 250 adults. On the survey the participants were asked to indicate how many hours of television, on average, they watch each week. They also completed The Perceived Threat Scale (PTS). The PTS produces a single score that indicates how much perceived threat one feels, with greater PTS scores indicating higher levels of perceived threat. Dr. Ortiz analyzed the relationship between hours of television watching with perceived threat and found a statistically significant effect (r (248) = +.75, p < .05). He communicated his findings to other psychologists at a regional meeting in a poster presentation. The title of the poster was "T.V. Causes People to Feel Threatened". Do you agree with Dr. Ortiz' conclusion about television and perceived threat? What do you think about his poster presentation? Fix it!

Fix It Lexi is a veterinary student who wants to know whether dog owners are happier than cat owners, and whether any difference in happiness between dog and cat owners also varies according to sex. She recruits volunteers by placing flyers at local veterinary clinics that indicate participants are needed for a study on happiness. After several weeks and only three volunteers, Lexi gets permission from the veterinary clinics to tell clients they must participate in the study in order for their pets to receive services. She also begins to give people a $25 credit toward veterinary services for their participation. Eventually Lexi gets 80 volunteers (40 men and 40 women). After giving informed consent, participants complete a demographic questionnaire, as well as the Subjective Happiness Scale. On the demographic questionnaire the participants reported their sex, age, ethnicity, how many pets they owned, what kind of pets they owned, and for how many years they owned each pet. Lexi used the demographic questionnaire to divide participants into four groups: men who own dogs (n = 23), men who own cats (n = 17), women who own dogs (n = 20), and women who own cats (n = 20). The mean happiness for each group was computed and compared using several t-tests. The results indicate men are happier than women, cat owners are happier than dog owners, men with dogs are happier than men with cats, women with cats are happier than women with dogs, men with cats are happier than women with dogs, and women with cats are happier than men with dogs. Lexi concludes that a relationship exists between pet ownership and happiness. What do you think of how Lexi obtained her participants? Do you agree with how she analyzed the data? Fix it!

Fix It A clinician was interested in understanding the effect of maternal age and maternal health on in vitro fertilization success. A total of 1078 women were identified as undergoing in vitro fertilization in a large metropolitan city. A random sample of 500 women was selected from this population and sent a letter requesting their participation in the study. A total of 108 women agreed to participate. A semi-structured interview was conducted with each woman. Interviews took between one and three hours to complete. Aside from asking basic demographic information such as age, ethnicity, and highest level of educational attainment, the interviewer asked questions to assess each woman's physical health. The final question of the survey was "How have you changed your diet and exercise routine to prepare for in vitro fertilization?" Data were analyzed using both quantitative and qualitative methods. Quantitative analyses included dividing respondents into three age groups: young (25-29 years), old (30-34 years), and older (35-39 years), and into three health categories based on BMI (normal, overweight and obese) at time of conception. The number of successful pregnancies across age groups and health level was measured. A successful pregnancy was defined as becoming pregnant and carrying the baby to term. This 3 × 3 design was analyzed statistically using a two-factor ANOVA. The data revealed a statistically significant main effect for age (p < .05), health level (p < .05), and an interaction for age × health (p < .05). Further inspection of the data suggested that younger women in the normal and overweight BMI range had the most successful pregnancies. Young women who were obese had a similar success rate as older normal-weight women. Qualitative analysis of how women prepared for in vitro revealed a statistical difference between older and younger women, with older women changing their diet and exercise routine much more than younger women. These results indicate that a woman's age and weight are factors associated with in vitro success, and that behavioral modifications, such as diet and exercise have little impact. What issues do you see with how the data were measured, analyzed and interpreted? Fix it!

Fix It A health psychologist is interested in understanding the effect of age of first menstrual cycle and fertility. A random sample of 300 women is selected from a total population of 803 female patients across three clinics in a small northeastern Texas city. The women in the sample are sent a letter requesting their participation in the study. A total of 82 women provide their informed consent to participate. The researcher contacts each volunteer via telephone and asks them the age at which they first menstruated, the total number of pregnancies they had, and the total number of live births they experienced. In addition, the researchers ask the women questions about their ethnicity, highest level of educational attainment, and household income. Quantitative analyses revealed that the earliest age of menstruation was 9 years, and the oldest age was 17 years. The researcher divides age of first menstruation into three categories: early (9-11), average (12-14), and late (15-17). She calculates the mean number of pregnancies for each age group. She also decides to determine if educational attainment affects fertility and compares the mean number of births by women with a high school diploma (or less), and the mean number of births by women with greater than a high school diploma. Her first analysis indicates that women who menstruate early in life are less fertile than those who menstruate at a typical age, or those who menstruate later in life. Her second analysis reveals that those with more education are more fertile than those with less education. The researcher concludes that while age of menstrual onset is a factor associated with fertility, the ability to become pregnant can be increased with education. Identify the strengths and limitations of this study. For each limitation offer a solution.

Fix It Brooke is a new graduate student. Her advisor, Dr. Martinez, has been studying the impact of experience on cognitive function for years. Dr. Martinez assigns Brooke several journal articles to read. She finds one in particular that piques her interest. The paper describes that rats that voluntarily exercise for 30 days perform better on the Morris Water Maze, a task indicative of declarative memory. She wonders whether this is true in humans, and whether the positive effect of exercise occurs sooner than 30 days following exercise. To that end, Brooke places an ad in the student newspaper to recruit volunteers. Brooke randomly selects 80 participants from the 128 people who responded to the ad. Participants are first assigned to an exercise condition or sedentary condition. Each condition is further divided so that some people participate for 2 weeks, some for 4 weeks, and others for 6 weeks. Participants in the exercise condition are asked to exercise at least twice a week. Participants are asked to keep a diary of their activity and also note their heart rate at five-minute intervals throughout their workout session. Sedentary participants are asked to not change their daily habits. After participating, each person returns to the lab to complete the TOMAL-2, a test to assess declarative memory (higher scores indicate better memory). The results revealed that participants who exercised for 4 weeks and those who exercised for 6 weeks performed much better on the TOMAL-2 than those who did not exercise at all. Interestingly, those who exercised for only 2 weeks performed similarly to those who did not exercise at all. These results suggest that a minimum of 4 weeks of exercise is needed to improve cognitive performance in adults. How would you describe this research design? What are its advantages? Discuss the operational definition of exercise and describe how and why you might change it.

Fix It Dr. Vorkman was asked by The Journal of Scientific Research to review an article. The article was sent to him via post and did not include the names of the authors. In the paper the authors describe a relationship between color and happiness. They specifically discuss how individuals who had listened to a choir of singers wearing yellow robes reported more happiness than those who had listed to a choir of singers wearing black robes. The study was conducted by recruiting a 25-member choir from a local church selected at random. The choir was given yellow robes to wear in the summer, and black robes to wear in the winter. To assess happiness, parishioners completed a Happiness Survey after a service in July and after a service in January. The choir performed at both services. A t-test to compare the median happiness scores in July and January was performed. The results indicated parishioners were significantly happier in July. The authors conclude color affects happiness and advocate for a ban on drab-colored clothing. What do you think about the article review process? What feedback do you think Dr. Vorkman should provide to The Journal of Scientific Research?

Fix It Lucy is a doing her senior thesis on how music makes people feel. She believes that the best approach to this topic is qualitative, since one's reaction to music is very personal and varies significantly from person to person. Lucy recruits 12 of her friends and family to participate in the study. She invites the participants to a focus group. Lucy decides to have a friend who is unfamiliar with the research topic moderate the focus group. Lucy's rationale is that she herself may be too biased and may influence her participants' responses; by having someone unfamiliar with the topic moderate, the discussion will be more authentic. After the focus group is finished the participants document their informed consent by signing their names to a sign-in sheet. Lucy transcribes the audio recordings the moderator made during the focus group and analyzes the data using t-tests and Pearson's r. She creates a poster to convey the study's results. On her poster she describes that women were significantly more likely than men to respond emotionally to music. What design and analysis problems exist in Lucy's study? What do you think about how she communicated the results? Fix it!

Fix It Dr. Cigainero hypothesizes that people who prefer horror movies are naturally less fearful than those who prefer romantic comedies. She also hypothesizes that fearfulness varies by sex. To test her hypothesis she asks male and female undergraduates what their favorite movie genre is. She randomly selects 25 men and 25 women who indicated horror, and 25 men and 25 women who indicated romantic comedy. These 100 participants are then invited to complete the 72-item Fear Survey Schedule (FSS). The higher someone scores on the FSS, the more generalized fear and anxiety that person has. The descriptive data indicate men and women who prefer horror movies reported less fear than men and women who prefer romantic comedies. The t-tests Dr. Cigainero used to analyze the data support the descriptive statistics. Men who preferred horror movies were significantly less fearful than men who preferred romantic comedies (t(49) = 4.72, p >.05), and women who preferred horror movies were significantly less fearful than women who preferred romantic comedies (t(49) = 3.89, p > .05). What was wrong with the statistical measures used in this study? Fix it!