In this exercise we analyze data that was collected as part of an experiment that investigated the mental relation between weapon and aggression.

In this exercise we analyze data that was collected as part of an experiment that investigated the mental relation between weapon and aggression. The experiment involved 32 undergraduate students ranging in age between 18 and 24 years. They were told that the purpose of this study was to test reading ability of various words. Each student was exposed to 192 trials. In each trial a computer presented a priming stimulus word for 1.25 seconds, a blank screen for 0.5 seconds, and then a target word. The experimenter instructed the subjects to read the first word to themselves and then to read the second word out loud as quickly as they could. The computer recorded reaction times.

The 192 trials were divided into 4 groups of size 48 each. In one group the priming stimulus word was a weapon word (such as "dagger" or "bullet") and the target word was an aggressive word (such as "destroy" or "wound"). In another group the target word was instead a nonaggressive word (such as "evacuate" or "ashamed"). The third and fourth groups involved a name of an animal (such as "rabbit" or "dog") as the priming stimulus word and the target, again, was either an aggressive or a nonaggressive word.

The file "weapons.csv" contains, for each subject and for each group of trials, the average of the 48 reaction times. The file can be found here (http://pluto.huji.ac.il/~msby/StatThink/Datasets/weapons.csv).

The file contains 5 variables:

• sex = The sex of the subject.
• aw = Aggressive target word following a weapon prime.
• an = Aggressive target word following a nonweapon (animal) prime.
• cw = Control target word (nonaggressive) following a weapon prime.
• cn = Control target word (nonaggressive) following a nonweapon prime.

-------------------------------------------------------------------------------------------------------------------------------------------------------

Question 1

Mark all the variables of type "factor":

Select one or more:

a. sex

b. aw

c. an

d. cw

e. cn

-------------------------------------------------------------------------------------------------------------------------------------------------------

Question 2

The total number of missing values is:

Answer:

-------------------------------------------------------------------------------------------------------------------------------------------------------

Question 3

Mark all the null hypotheses that are not rejected with a significance level of 5%:

Select one or more:

a. The expected value of "an" is the same both for males and for females.

b. The expected value of "aw" is the same both for males and for females.

c. The expected value of "cn" is the same both for males and for females.

d. The expected value of "cw" is the same both for males and for females.

e. All of the given null hypotheses are rejected.

-------------------------------------------------------------------------------------------------------------------------------------------------------

Question 4

Mark all the null hypotheses that are not rejected with a significance level of 5%:

Select one or more:

a. The variance of "an" is the same both for males and for females.

b. The variance of "aw" is the same both for males and for females.

c. The variance of "cn" is the same both for males and for females.

d. The variance of "cw" is the same both for males and for females.

e. All of the given null hypotheses are rejected.

-------------------------------------------------------------------------------------------------------------------------------------------------------

Question 5

One may measure the effect of a "weapon" primer on the speed in naming a word associated with aggression by considering the difference "aw - an". Call this difference "da". Likewise, one may measure the effect of a "weapon" primer on the speed in naming a neutral word by considering the difference "cw - cn". Call this other difference "dc". An overall contrast that may be used in order to test the relation between a "weapon" primer and a target associated with aggression is the difference "da - dc". Apply the t-test in test the null hypothesis that the expectation of the contrast is zero against a two-sided alternative. The resulting p-value is: (Give an answer of the form x.xxx, i.e. with 3 significance digits)

Answer:

-------------------------------------------------------------------------------------------------------------------------------------------------------

Question 6

Construct a 95%-confidence for "aw - an" and a 95%-confidence interval for "cw - cn". The value 0 is contained in:

Select one:

a. Only in the confidence interval for "aw - an".

b. Only in the confidence interval for "cw - cn".

c. Both confidence intervals.

d. Neither of the confidence intervals.

-------------------------------------------------------------------------------------------------------------------------------------------------------

Question 7

The variable "aw" is the reaction time for a target word associated with aggression when the primer is a word associated with weapons. The variability of this variable between the subjects may be attributed in part to variability in reading ability among people. One may propose the variable "an" as measure of reading ability of words associated with aggression when the "weapon" effect is not present. Let "an" be an explanatory variable and let "aw" be the response. The percentage of the variance that is explained by the explanatory variable is:

Select one:

a. Less than 25%.

b. Between 25% and 50%.

c. Between 50% and 75%.

d. More than 75%.

-------------------------------------------------------------------------------------------------------------------------------------------------------

Question 8

Consider a regression model where "an" is the explanatory variable and "aw" is the response. Consider the 95%-confidence intervals for the intercept and the slope. Then:

Select one:

a. The value 0 belongs to the confidence interval for the intercept and the value 1 belongs to the confidence interval for the slope.

b. The value 0 belongs to the confidence interval for the intercept but the value 1 does not belong to the confidence interval for the slope.

c. The value 0 does not belongs to the confidence interval for the intercept but the value 1 belong to the confidence interval for the slope.

d. The value 0 does not belong to the confidence interval for the intercept and the value 1 does not belong to the confidence interval for the slope.