Question 1

In the 2000 presidential election, the polls of upper-middle-class Americans tended to show Bush leading Gore by a landslide. When election night came, the final results were closer than had been predicted, and it turned out that Gore won the popular vote while Bush won the more critical Electoral College vote. One possible problem with the earlier survey results was that they relied primarily on polls of upper-middle-class people. This illustrates the finding that a major problem with surveys is

Accepted Answer

The given scenario shows that the survey results were not representative of the population, which resulted in incorrect predictions. Therefore, a major problem with surveys is that they are not predictive of behavior if the sample is not representative of the population about which the prediction is to be made.

Question 2

A recent study found that the more time fathers spend with their children, the less likely they are to abuse their children. Which of the following is the most valid interpretation of this finding?

Accepted Answer

The answer of A recent study found that the more...

Question 3

Recall the description of the political poll taken by Literary Digest about the Landon vs. Roosevelt election. This is an example of a potential error that can be made when doing survey research involving political polls. What is the lesson that survey researchers learned from this error?

Accepted Answer

The lesson learned from the Literary Digest poll is that it is important to use random selection to obtain a representative sample. The poll used only magazine subscribers and automobile owners as their sample, which skewed the results and led to a significant error in predicting the election outcome. Using random selection helps ensure that everyone in the population has an equal chance of being included in the sample, leading to more accurate results.

Question 4

Imagine that researchers have found a positive correlation between the frequency of disagreements that couples have and how long they stay together. Based on this correlation, would you start arguments with your significant other in order to sustain your relationship?

Accepted Answer

It is important to recognize that although a positive correlation has been found, it does not mean that causing arguments will actually help sustain a relationship. Correlation does not prove causation. It is also important to consider other factors that may be influencing the relationship, such as communication skills and overall compatibility. Therefore, it is not advisable to start arguments with your significant other solely based on the findings of the correlation.

Question 5

In order to obtain the most accurate responses, questions on a survey should be

Accepted Answer

Straightforward questions are clear and specific, making it easier for participants to understand and respond accurately. Brevity can sometimes sacrifice the necessary level of detail for accurate responses. Complex questions may confuse or overwhelm participants, leading to inconsistent or inaccurate answers. Carefully wording questions to detect lying is unethical and could bias results.

Question 6

Based on survey data, the Literary Digest erroneously predicted that Alf Landon would beat Franklin D. Roosevelt in the upcoming presidential election. What cardinal rule of survey methods did the Literary Digest violate? Make sure

Accepted Answer

The Literary Digest violated the cardinal rule of survey methods by not sampling randomly from the population. They relied on their own subscriber lists and telephone directories, which produced a biased sample that over-represented affluent individuals and Republicans. Random sampling is necessary to ensure that survey results are representative of the population as a whole.

Question 7

Consider the following survey item: "If you found yourself on an airplane with engine problems, would you ________?" Most social psychologists would not include such an item on their survey because most respondents would

Accepted Answer

Most respondents would find it difficult to accurately imagine what they would actually do in such a high-stress situation, thus making their responses unreliable.

Question 8

A researcher wants to ask hypothetical questions or for explanations of past behavior on a survey. Based on what the authors of your text report about survey questions, this researcher should consider asking

Accepted Answer

Straightforward questions are generally more reliable and easier for respondents to answer accurately, as hypothetical questions or those asking for explanations can lead to speculation and less accurate responses.

Question 9

After reading Chapter 2 (Methodology), what would you say to a president who proudly takes credit for the following events happening during his term: lowering the deficit, reducing crime, and increasing literacy?

Accepted Answer

The president's claim may show correlation but it does not prove causation. Without proper research and analysis, it is impossible to determine whether the president's actions directly caused the observed outcomes.

Question 10

A researcher discovers a strong positive correlation between ice cream sales and the incidence of violent crimes. Based on this, we can conclude that

Accepted Answer

The correlation only shows that on days when ice cream sales are high, there are likely to be more violent crimes. It does not necessarily mean that eating ice cream causes people to become violent or that there is a causal relationship between the two variables. It also does not suggest that cutbacks in the sales of ice cream or controlling crime rates will have a direct impact on each other.

Question 11

Which of the following best defines the term "sample"?

Accepted Answer

A sample refers to the people who are actually measured in a study, as opposed to the entire population or a subset of the population that researchers hope to generalize the results to.

Question 12

There is a moderately strong correlation between the number of bars in a town and the number of churches in a town. As it turns out, this correlation is due to the correlation of both numbers of bars and number of churches with town population. This illustrates which of the following? Correlations are

Accepted Answer

Correlation does not necessarily imply causation. The fact that both the number of bars and the number of churches are correlated with town population suggests that population size is the common factor driving the correlation between the number of bars and the number of churches. Therefore, there is no direct or causal relationship between bars and churches in the town.

Question 13

Carlos, the president of a fraternity on campus, randomly sampled 500 students, asked them if they belonged to a fraternity or sorority, and asked them about their current GPA. He discovered that the GPAs of those people in fraternities and sororities are higher than those of people who are not involved in the Greek system. Gleefully, he presented his findings to the dean, saying that being involved in a fraternity or sorority leads to higher grades. What rule of research methods is Carlos breaking?

Accepted Answer

Carlos is breaking the rule of correlation does not equal causation. Just because there is a correlation between involvement in Greek life and higher GPAs, it does not mean that being in a fraternity or sorority actually causes higher grades. It is possible that there are other factors at play, such as the type of students who join fraternities and sororities, or the resources available to Greek organizations, that lead to higher grades.

Question 14

Based on the idea of "telling more than you can know" from your text, which of the following survey questions would be least likely to yield accurate responses?

Accepted Answer

The answer of Based on the idea of "telling more...

Question 15

Using random selection is a way to ensure that

Accepted Answer

Random selection ensures that every member of a population has an equal chance of being selected for a sample, which in turn increases the likelihood that the sample will be representative of the population. This is important for drawing accurate conclusions and making generalizations about the population.

Question 16

Based on Nisbett and Wilson's (1977) work (as presented by the authors of your text) on telling more than you can know, if a survey respondent is asked about a hypothetical situation, he or she is probably going to respond

Accepted Answer

According to Nisbett and Wilson's work, people often rely on their theories and beliefs about what should have influenced them when asked about a hypothetical situation, rather than their actual thoughts and behaviors. This is known as the "introspection illusion." Therefore, choice A is the best answer as it reflects this finding.

Question 17

Rosenberg and his colleagues (1992) conducted a study that found that women who relied on the diaphragm or contraceptive sponge had fewer STDs than women who used condoms. The media jumped to the conclusion that condom use contributes to STDs. Of the limitations of this study, which one should have prevented the media from drawing a causal conclusion?

Accepted Answer

The limitation that the choice to use condoms may have stemmed from STD contraction rather than the reverse should have prevented the media from drawing a causal conclusion. This means that women who were already at a higher risk for STDs may have been more likely to choose to use condoms, rather than the condoms themselves causing the increase in STDs.

Question 18

Based on the error in sampling in political polls that was described in your text, how would you recommend obtaining a representative sample for a poll (survey) about political candidates?

Accepted Answer

The best choice is to give the poll face-to-face at an average supermarket or post office. This method will allow for a more random selection of participants and will give a more representative sample compared to conducting phone surveys using home phone numbers, which would only reach a certain demographic. Polling people at the local office which provides Food Stamps and other forms of assistance would not give a representative sample as it would only target a certain population. Similarly, giving the poll face-to-face at college campuses nationwide would only target a specific age group and socioeconomic status.

Question 19

Why is it unwise to conclude that if two variables are correlated, one must have caused the other?

Accepted Answer

Correlation does not necessarily imply causation because there may be a third variable influencing the relationship between the two variables being studied. This is known as a confounding variable. Thus, it is unwise to conclude that one variable causes the other solely based on correlation.

Question 20

The idea of "telling more than you can know" (Nisbett & Wilson, 1977) is the premise that people may report why they respond a certain way, but that their reports about causes of behavior reflect

Accepted Answer

According to Nisbett and Wilson (1977), people often rely on their theories and beliefs about what should have influenced them to explain their behavior instead of accurately reporting the true causes. Therefore, choice B is the correct answer. Options A, C, and D do not fit with the premise of the "telling more than you can know" phenomenon.

Quiz 2: Methodology: How Social Psychologists Do Research