POS 401 Quiz 1

docx

School

Arizona State University *

*We aren’t endorsed by this school

Course

401

Subject

Statistics

Date

Apr 3, 2024

Type

docx

Pages

6

Uploaded by ChancellorPheasantPerson390

Report
Score for this attempt: 9 out of 9 Submitted Mar 15 at 10:40pm This attempt took 2 minutes. Question 1 1 / 1 pts What are some of the reasons why it is difficult to gain knowledge about politics? Correct! Political actors do not always act in the expected ways. This is the individuality problem. Our understanding of human behavior is certainly incomplete, and even it was not, humans are not automatons. This makes predicting the behavior of individuals a very hard thing to do. However, we can still do a good job at predicting what the behavior will be *on average*. Statistics can tell us how confident we should be about these abstract predictions. The interpretation of political events is always subjective We do not have enough data on politics. Correct! Political events are the result of many interacting forces. This is what we called the complexity problem. Unlike in other scientific disciplines, the experimental method has limited applicability in politics. This means that is often hard to isolate the effect of any one factor in the overall outcome we are interested in. Do social movements succeed when popular dissatisfaction is high, when the government is too weak to resist them, or when the international community supports them? It may be all three, but it's hard to tell because we cannot change the data to fit our needs as you would do in an experiment. Moreover, if it's all three, which one is more important? Statistics can help us here by quantifying their respective influence.
Question 2 1 / 1 pts Which of the following is a conclusion of the reliabilist view of science? Studies that cannot explain why something happened are unscientific Correct! Science is characterized by its method Recall reliabilism argues that we cannot know if a belief is true and justified until tested. We can only know if the process that we followed to obtain that belief produces "good" answers (i.e. true and justified) most of the time. Therefore, it suggests that we should focus on the methods to assess scientific quality. Research is scientific when it allows us to predict what will happen in the future Question 3 1 / 1 pts The following may is a reason why our statistical analysis may produce a wrong answer Lack of data Correct! Inadequate research design
Statistics tells us what to except given the evidence. However, if the evidence we are analyzing is not the right one, statistics cannot give us the right answer. An example of this is survivor bias, where we analyze an incomplete and non-representative sample to learn about the entire population of cases. Correct! Inadequate measurement Statistics depend on the quality of our evidence. If our evidence misrepresents what is out in the world because of measurement error, then the statistical analysis will be wrong. While statistics have answers to the problem of measurement error, researchers need to take explicit steps to address it. Otherwise, we are left with bad evidence and bad inferences. Question 4 1 / 1 pts How do we define statistics? Statistics is the application of mathematics to uncover patterns in observed data. Statistics is the application of mathematics to make predictions about unobserved data. Statistics is a way to summarize complex data. Correct! The application of mathematics to uncover patterns in observed data and make probabilistic predictions about unobserved data. This definition highlights three important elements: the mathematical backing of statistics, the differentiation between observed (known) data and unobserved (unknown) data, and the generation of predictions with quantified uncertainty.
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
Question 5 1 / 1 pts Statistical predictions are based on identifying the event with 'maximum likelihood'. How is this estimated? Through combinatorics Correct! We count the number of ways in which each outcome could have occurred and choose the one that can emerge in the greatest number of ways. There are many ways in which we can describe the inner workings of statistics. This is the traditional Bayesian description; you might find others elsewhere. Question 6 1 / 1 pts What do we mean when we say a research question must be motivated by theory? It means we should be concerned with abstract categories It refers to previous work in the area Correct! It must explain why things occur. Take for example the difference between saying “inflation has increased in the last year” to “inflations has increased in the last year due to problems with the supply chain”. The latter may be true, but it does not tell us
why inflation has occurred. Good statistical work would show that supply chain issues increase inflation, thus explaining why inflation has occurred. Question 7 1 / 1 pts What do we mean when we say that a hypothesis should be falsifiable? That we will gather evidence to show the hypothesis is wrong. Correct! That some evidence could show that the hypothesis is wrong. The textbook refers to this as a hypothesis (or research question) being answerable. Question 8 1 / 1 pts What is the problem of observational equivalence? Correct! That two competing theories have indistinguishable observable implications This is bad news because then we will not have data to judge which one of these two theories does a better job at explaining reality. This is a problem different from falsification, however. In the latter, no possible evidence can show the theory is wrong. In observational equivalence, we cannot show that the evidence supports one theory over the other. That our observations are very similar to each other
Question 9 1 / 1 pts In a dataset, we have variables and observations. Which one is which? Variables are the columns in our dataset Correct! Observations are our unit of analysis. They are the cases that we are studying, and for which we gather the necessary data to test our theory. Correct! Variables are elements that can be measured or categorized that contain information about our cases. Observations are the rows in our dataset. Quiz Score: 9 out of 9 Previous Next
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help