Data are skewed right. Which is larger, the mean or the median?
What is the mean?
What does a z-score of −1.5 mean?
The value is 1.5 standard deviations below the mean.
Number every student 1 to 500. A random number generator picks 50 numbers. What type of sample is this?
SRS
A study tests 3 fertilizers and measures plant height. Name the explanatory variable and the response variable.
Explanatory: type of fertilizer. Response: plant height.
Which r shows the strongest linear relationship: 0.3, −0.9, or 0.6?
−0.9. The strength depends on how close r is to 1 or −1.
Data: 4, 6, 7, 9, 14. Find the median and the range.
Median = 7. Range = 14 − 4 = 10.
Mean = 70, SD = 10. Find the z-score for a score of 85.
z = (85 − 70) / 10 = 1.5
Students are split by grade (9, 10, 11, 12) and 10 are chosen at random from each grade. Stratified or cluster? Why?
Stratified. Students are chosen from every group. A cluster sample would pick whole groups and survey everyone in them.
Why do researchers block before they randomly assign subjects to treatment groups?
It makes the groups similar, so differences can be linked to the treatment.
ŷ = 12 + 3x, where x = hours studied and y = test score. Interpret the slope.
For each extra hour studied, the predicted score increases by 3 points.
Q1 = 20 and Q3 = 32. Find the IQR and the upper cutoff for outliers.
IQR = 12. Upper cutoff = 32 + 1.5(12) = 50.
Mean = 100, SD = 15. About what percent of values are between 70 and 130?
About 95% (within 2 standard deviations).
A survey is mailed to 1,000 people. Only 15% answer, and they mostly have strong opinions. What kind of bias is this?
Nonresponse
What is blocking, and why is it used?
Group similar subjects into blocks, then randomize within each block. It reduces variability.
Use ŷ = 12 + 3x. Predict y when x = 5. If the actual y is 30, what is the residual?
Predicted = 27. Residual = 30 − 27 = 3
Every value in a data set increases by 5. What happens to the mean, the standard deviation, and the IQR?
The mean increases by 5. The standard deviation and IQR stay the same.
Quiz A: score 92, mean 80, SD 6. Quiz B: score 75, mean 60, SD 10. Which score is better compared to its class?
Quiz A. z = 2.0 for A and z = 1.5 for B.
A survey asks, “Don’t you agree the school day is too long?” What kind of bias is this?
Question wording bias
100 students who chose to nap are compared with 100 who did not, on test scores. Experiment or observational? Can we say naps cause higher scores?
Observational. No. Other (confounding) variables could explain the difference.
r² = 0.64. What does it mean?
64% of the variation in y is explained by its linear relationship with x.