sta 210 exam
Confounding often defeats attempts to show that one variable causes changes in another variable. Confounding means that a. the effects of two or more variables are mixed up, so we cannot say which is causing the response. b. we don't know which is the response variable and which is the explanatory variable. c. we would get widely varied results if we repeated the study many times. d. we have scheduled too many treatments for the number of subjects we have.
A
Which of the following statements do you think could possibly be true? a. The number of students enrolled at Midville University decreased by 10.4% last year. b. A basketball team took 20 free throws in a game last week and made 72.6% of them. c. Yesterday it was 30° (Fahrenheit) in Chicago. Today it warmed up to 60°. This is a 50% increase in the temperature. d. My weight decreased by 10% last year but then increased by 10% in the first two months of this year. Thus, my overall weight from the beginning of last year until now is unchanged.
A
Which one of these statistics is affected by outliers? a. Mean b. Interquartile range c. Standard deviation d. Range
A
indicate whether the statement implies causation or just indicates association without causation. Aging of the brain tends to be delayed in people with a college education.
Association without causation. The statement is not claiming that a college education will cause a delay in aging of the brain. In fact, there are many confounding factors.
Explain how slippery evidence affects human inference?
At first glance we infer one thing, but with further review we see something else is true
Have a look at the following table where we have classified all corrective lens wearers by sex and type. Male FemaleEyeglasses only (millions) 43.3 54.6Contact lenses (millions) 2.9 7.0TOTAL 46.2 61.6 From the table, 11.4% of female corrective lens wearers wear contact lenses. The corresponding percentage for males is a. 2.9%. b. 6.3%. c. 6.7%. d. None of the above.
B
Which of the following statements is true? a. Legitimate correlation never implies causation. b. Legitimate correlation does not necessarily imply causation. c. Legitimate correlation is equivalent to causation.
B
Which of these is an example of Simpson's paradox? a. Teachers' salaries and sales of alcoholic beverages have risen together over time, but paying teachers more does not cause higher alcohol sales. b. In the 1970s males were admitted to each of six departments at Berkeley at rates at essentially equal to or less than females, but when the data from all six departments were combined, males showed a much higher overall admission rate. c. The percent of surgery patients given Anesthetic A who die is higher than the percent for Anesthetic B, but this is because A is used in more serious surgeries. d. States in which a smaller percent of students take the SAT exam have higher median scores on the SAT.
B
A student reports that of a random sample of 200 college undergraduates, 37% were female and 41% were male. We know that the student has made a mistake because a. it is highly unlikely that a sample of 200 undergraduates would have as few as 37% females. b. we know from class that approximately 60% of all undergraduates are female c. 37% plus 41% is not 100%. d. 37% of 200 would mean there was a "partial female" in the sample.
C
In the Nova Southeastern on-line learning example versus face -to face discussed in class, the primary source of confounding was caused by (have a look at BN 1.14) a. massive, unexpected computer failures for the on-line group b. the placebo effect c. one group of subjects having better prepared backgrounds than the other d. the randomization of students to the two treatments
C
What is the effect of an outlier on the value of a correlation coefficient? a. An outlier will always decrease a correlation coefficient. b. An outlier will always increase a correlation coefficient. c. An outlier might either decrease or increase a correlation coefficient, depending on where it is in relation to the other points. d. An outlier will have no effect on a correlation coefficient.
C
Over 30,000 people participated in an online poll on cnn.com conducted in April 2012 asking "Have you ever driven with a pet on your lap?" We see that 34% of the participants answered yes and 66% answered now. Is the variable in this study quantitative or categorical?
Categorical
indicate whether the statement implies causation or just indicates association without causation. If you study more, your grade will improve in this course.
Causation! "Will improve" implies causation.
indicate whether the statement implies causation or just indicates association without causation. Want to lose weight? Drink tea!
Causation! The implication is that drinking tea will cause you to lose weight.
Explain the difference between Correlation and Causation.
Correlation is a straight line association between two variables. It is visual too. Causation is when one variable/thing causes the other.
**Suppose a correlation is found to be very weak. What does this mean about the relationship between the two variables? a. There is no linear relationship between the two variables being measured. b. There may be separate linear relationships that are being masked by a third variable that was not accounted for. c. There may be a different type of relationship between the variables; just not a linear one. d. All of the above.
D
Dr. Stat's library of literary classics includes issues of certain not-so-scholarly periodicals. The total number of issues of four such periodicals, together with the percentages of Dr. Stat's entire library that they comprise, are as follows: Solder of World Wrestling Periodical Spider-Man Fortune Federation Mad Number 156 52 26 78 % of Library 18.6% 6.2% 3.1% 7.0% One of the percentages in this table is incorrect. Which is it? a. 18.6% (Spider-Man) b. 6.2% (Soldier of Fortune) c. 3.1% (WWF) d. 7.0% (Mad)
D
How well does the number of beers a student drinks predict his or her blood alcohol content? Sixteen student volunteers at The Ohio State University drank a randomly assigned number of cans of beer. Thirty minutes later, a police officer measured their blood alcohol content (BAC). A scatterplot of the data appears below. The scatterplot shows a. a weak negative relationship. b. a moderately strong negative relationship. c. almost no relationship. d. a moderately strong positive straight-line relationship between number of beers and BAC.
D
Response Variable:
Dependent Variable
In each case below, we describe a study to determine whether exercise helps increase certain mood enhancing chemicals in the brain. Indicate whether each describes an experiment or an observational study. Using a random sample of 100 people, we randomly assign half of them to participate in a regular exercise program for a six-week period while the other half makes no changes. At the end of the time period, we measure the brain chemicals.
Experiment. The experimenters actively imposed the explanatory variable (exercise or not).
Suppose the correlation between height and weight for all the students in your class is found to be r=0.20. You think that you must be overlooking something, because the relationship should be stronger than that. What additional variable may be masking an underlying relationship here?
Gender and age
Explanatory Variable
Independent Variable
Explain how (if) an outlier can affect the correlation of a data set.
It will change it
What is the link between a lurking variable and Confounding?
Lurking variable causes confounding
How are lurking variables related to Simpson's Paradox?
Lurking variables can cause Simpson's Paradox. We need to split the table up by The lurking variable.
In each case, do you expect a positive or negative association between the two quantitative variables? Age and maximum running speed, for adults
Negative: Older people (such as 80-year-olds) tend to run slower than younger people (such as 20-year-olds).
Researchers noticed that happier heart patients are much more likely to still be alive 10 years down the road than unhappy heart patients. Does this mean that if an unhappy heart patient suddenly decides to start being happy, this will cause them to live longer?
No, this is an observational study, not a controlled experiment. Many confounding variables exist. For example exercise.
In each case below, we describe a study to determine whether exercise helps increase certain mood enhancing chemicals in the brain. Indicate whether each describes an experiment or an observational study. We contact a random sample of 100 people and record how much each person exercises and also measure the chemicals in the brain for each person.
Observational study. No variables were manipulated.
In each case, do you expect a positive or negative association between the two quantitative variables? Age of the husband and age of the wife, for married couples
Positive: 80-year-olds are more likely to be married to other 80-year-olds than to 20-year-olds. Ask the students what a negative relationship between these two variables would mean!
In each case, do you expect a positive or negative association between the two quantitative variables? Age and maximum running speed, for children.
Positive: Older children (such as 12-year-olds) tend to run faster than younger children (such as 4-year-olds).
In each case, do you expect a positive or negative association between the two quantitative variables? Number of years of education and annual salary, for US adults
Positive: people with more education generally make higher salaries.
Princeton Survey Research reports on a survey of 1,917 cell phone users in the US, conducted in May 2010, asking "On an average day, about how many phone calls do you make and receive on your cell phone?" Is the variable in this study quantitative or categorical?
Quantitative
Why, in the absence of any other evidence, can't you use data from an observational study to establish a causal link between two measurement variables?
The subjects are not randomly assigned to treatments. There could be confounding variables.
Why is it important to critically read news articles, and why should we be careful about spreading information we are not confident about?
We need to critically read to make sure that what we are reading is true
Placebo effect
a beneficial effect, produced by a placebo drug or treatment, that cannot be attributed to the properties of the placebo itself, and must therefore be due to the patient's belief in that treatment.
Correlation Coefficient:
a number between −1 and +1 calculated so as to represent the linear dependence of two variables or sets of data.
An association is described. In each case, indicate a). Whether the statement implies causation or just association b). A possible confounding variable Increased weight helps students run faster.
a). Causation b). Gender! This is an often overlooked confounding variable. Males tend to weigh more and also to run faster. Another possible answer is age if the participants are children.
An association is described. In each case, indicate a). Whether the statement implies causation or just association b). A possible confounding variable "Exercise reduces risk of Alzheimers" claims a headline reporting on a study of elderly people that recorded how much each exercised at age 70 and then whether the person got Alzheimer's disease.
a). Causation b). The person's overall health at age 70. People with early stage (and not yet detected) Alzheimer's disease are probably less likely to exercise.
An association is described. In each case, indicate a). Whether the statement implies causation or just association b). A possible confounding variable More sales of sunscreen tend to occur when more sunglasses are sold.
a). Just association without causation. b). Amount of sunshine
List three types of Slippery Evidence we talked about in the class. For each of the types of slippery evidence give an example explain how we can spot them. Recall all examples at the beginning of the semester that we did in class. Answers will be different.
a. Misplaced decimal point b. plots that have incorrect proportions (pictograms) c. Reporting relative percent's instead of absolute percent's
A data point that is far removed from the rest of the data is called a(n)_
outlier
