Question 1

1) Which of the following proposed relationships would be a negative correlation?&#10;A) As IQ increases so does memory span&#10;B) Reaction times are shorter the more caffeine participants are given&#10;C) Extroversion scores are higher when self-esteem is greater&#10;D) Longer television watching is related to more aggressive behaviour

Accepted Answer

Negative correlation means as one variable increases, the other decreases. In option B, as caffeine intake increases, reaction times are shorter (decrease).

Question 2

2) A researcher asks participants to run around a running track and then to take a sports events memory test. She records the position they came in the running and correlates these values with the scores obtained on the memory task, hypothesising that faster runners have better sports event memories.
What would be the appropriate test of correlation that she should use on her data?

A) Pearson's product moment correlation
B) Spearman's rank correlation
C) The point biserial correlation
D) The phi coefficient

B

Accepted Answer

Since the researcher is comparing the rank order of running position with the scores obtained on the memory test, the appropriate correlation test would be Spearman's rank correlation.

Question 3

3) If a correlation of zero is found between two variables this could mean that:&#10;A) There is a U shaped relationship between the variables&#10;B) The scores on one variable are the same for every participant&#10;C) There is no relationship between the variables&#10;D) All of the above

Accepted Answer

A correlation of zero can indicate a non-linear relationship (such as a U-shaped relationship), that one variable does not vary, or genuinely no relationship between the variables.

Question 4

Consider the following data set gathered from 16 year olds where 'violent TV' is the number of hours of aggressive television the teenager reported viewing per week and 'aggression' is their score on an aggression scale where a high score indicates a high level of aggression.
$\begin{array} { | c | c | } \hline \text { Violent TV } & \text { Aggression } \\\hline 15 & 53 \\\hline 28 & 52 \\\hline 23 & 56 \\\hline 21 & 65 \\\hline 14 & 27 \\\hline 31 & 83 \\\hline 24 & 65 \\\hline 19 & 58 \\\hline 35 & 83 \\\hline\end{array}$

-6) With the correlation data from above, with $\alpha$ at .05, would be about:

A) .62
B) 2.27
C) .805
D) .81

Accepted Answer

To calculate Pearson correlation, we can use a calculator or software. With a calculator, we can use the following formula:
r = (nΣXY - ΣXΣY) / [(nΣX^2 - (ΣX)^2)(nΣY^2 - (ΣY)^2)]^0.5
where n is the sample size, Σ represents the sum, X and Y represent the two variables, and ^0.5 represents the square root.
With the data given above, we can calculate:

n = 9
ΣX = 190
ΣY = 522
ΣX^2 = 5722
ΣY^2 = 35229
ΣXY = 11840

Substituting these values into the above formula, we get:
r = (9(11840) - 190(522)) / [(9(5722) - 190^2)(9(35229) - 522^2)]^0.5
r = 0.805

To determine the p-value, we need to consult a table of critical values of correlation coefficients at a given level of significance and degrees of freedom. With a sample size of 9 and a significance level of 0.01, we have 7 degrees of freedom. Consulting the table, we find that the critical value is 0.725. Since our calculated r value (0.805) is greater than the critical value, we can reject the null hypothesis that the correlation coefficient is zero at the 0.01 level of significance.

Therefore, the best choice is B, which reports the Pearson correlation result as r = 0.805 with p < 0.01.

Question 5

Consider the following data set gathered from 16 year olds where 'violent TV' is the number of hours of aggressive television the teenager reported viewing per week and 'aggression' is their score on an aggression scale where a high score indicates a high level of aggression.
$\begin{array} { | c | c | } \hline \text { Violent TV } & \text { Aggression } \\\hline 15 & 53 \\\hline 28 & 52 \\\hline 23 & 56 \\\hline 21 & 65 \\\hline 14 & 27 \\\hline 31 & 83 \\\hline 24 & 65 \\\hline 19 & 58 \\\hline 35 & 83 \\\hline\end{array}$

-6) With the correlation data from above, with $\alpha$ at .05, would be about:

A) .62
B) 2.27
C) .805
D) .81

Accepted Answer

The data can be used to argue that aggressive teenagers choose to watch more violent television because it shows a correlation between the number of hours of violent TV watched and aggression scores. However, correlation does not imply causation, so it cannot definitively prove that violent TV causes aggression or vice versa, but it can suggest a relationship where those with higher aggression levels might be more inclined to watch violent TV.

Question 6

Consider the following data set gathered from 16 year olds where 'violent TV' is the number of hours of aggressive television the teenager reported viewing per week and 'aggression' is their score on an aggression scale where a high score indicates a high level of aggression.
$\begin{array} { | c | c | } \hline \text { Violent TV } & \text { Aggression } \\\hline 15 & 53 \\\hline 28 & 52 \\\hline 23 & 56 \\\hline 21 & 65 \\\hline 14 & 27 \\\hline 31 & 83 \\\hline 24 & 65 \\\hline 19 & 58 \\\hline 35 & 83 \\\hline\end{array}$

-6) With the correlation data from above, with $\alpha$ at .05, would be about:

A) .62
B) 2.27
C) .805
D) .81

فتح الحزمة

افتح القفل للوصول البطاقات البالغ عددها 7 في هذه المجموعة.

فتح الحزمة

k this deck

Accepted Answer

The answer of Consider the following data set gathered from...

Question 7

7) Which of the following could not be a valid correlation coefficient?&#10;A) .76&#10;B) -.43&#10;C) 0&#10;D) 1.49

Accepted Answer

The correlation coefficient can range from -1 to +1. Therefore, the only invalid option is 1.49, as it falls outside of this range.

Deck 19: Correlation