October 15, 2014
(1) Systematic variation is
- variation caused by random effects in any direction
- variation coming from a factor that introduces bias in one direction only
- always inflating the standard deviation
- always deflating the standard deviation
- all of the above
(2) The x- and y-axes of a histogram represent…
- frequencies
- counts
- any metrics
- none of the above
(3) The standard error…
- equals the standard deviation divided by the square root of the sample size
- estimates the spread in the sample
- equals the mean in most cases
- often equals the standard deviation
- more than one in the above are correct
(4) Pick the variable that is categorical and nominal
- Rank in a 100 m race
- Civil status
- Numbers of cigarettes smoked per day
- Cholesterol level in blood samples (in mg/ml)
(5) Which of the following is not correct:
- An independent variable is the same as a predictor variable
- A predictor variable can be a factor
- A response variable is the same as a factor
- A factor can have two or more factor levels
- A response variable can be categorical or continuous
(6) Which operation in calculating the variance helps to make sure the latter does not automatically increase with sample size?
- Squaring the sum of the differences between observations and the mean
- Dividing by the degrees of freedom
- Summing up the differences between observations and the mean
- All of the above is correct
(7) In this sample: (1, 3, 7, 11, 15, 19, 20, 22, 24, 30, 31, 35, 39), …
- the median is 20
- the first quartile is 11
- the third quartile is 30
- the interquartile range is 21
- all of the above are correct
- only (1) is correct
(8) A population is known to be chi-square distributed with 4 degrees of freedom. What is the chance of finding a value greater than 10?
- 95%
- 50%
- 4%
- less than 4%
- Almost zero
(9) You want to find out whether protein content in yogurt (low, medium, high) affects the viscosity of the product. What test would you use?
- A two-tailed Wilcoxon test
- A two tailed t-test
- A correlation analysis
- An ANOVA approach
- A regression approach
(10) In a power t-test, if you increase your sample size AND you decrease your standard deviation,…
- you increase your power
- you increase your type I error probability
- you decrease your type II error probability
- you improve your chances to find a difference, should there be one
- All but (2) are correct
(11) After regression analysis, we need to check…
- whether the residuals are continuous
- whether the residuals are homogenous
- whether the residuals are normal
- all of the above
- only (2) and (3) are correct
(12) Which is true for the F-value in ANOVA?
- It represents a variance ratio
- The higher it is, the lower the associated p-value
- It is the between divided by the within-group variance
- It depends on the degrees of freedom (sample size within groups, number of groups)
- All of the above is correct
(13) Just by sketching a standard normal distribution, what is the probability of sampling a value greater than 2?
- the same as sampling a value lower than -2
- quite low, maybe a few percent
- around 30%
- infinitely small
- 1 and 2 are correct
(14) A confidence interval tells us…
- the probability with which the mean falls in a certain range
- the range within which the true mean falls in 95% of cases
- the range within which the true mean sits with 95% confidence
- none of the above
(15) In regression analysis, we try to…
- minimise the sum of squared differences between the observed values and the mean
- minimise the distances between the fitted values and the observed values
- minimise the sum of squared differences between the fitted values and the observed values
- maximise the sum of squared differences between the fitted values and the observed values