You are on page 1of 5

Hayley Hamilton and Rabab Zahidi

Fundamentals of Biostatistics Team Presentation


April 12th, 2016

Correlation and Regression Review Questions


Level 1:
1. Define correlation and give an example of a correlation.

a. Correlation is assessing the strength of the relationship between two variables. An


example of correlation is age and blood pressure.
a. Dependent variables are also known as? (Name two)
a. Outcome variables, response variables, or criterion variables.
a. In the linear function (y= + x), what does represent, what does it mean?
a. represents the y-intercept. Y-intercept is the value of y when x is zero.
a. Define residual in words and using symbols.
a. For an observation, the difference between an observed value and the predicted
value of the response variable is called residual. Y-
a. If the direction of the correlation coefficient (r) is positive, what does it mean?
What is another name for this type of correlation?
a. If the direction of the correlation coefficient is positive this means that high scores on
one variable are associated with high scores on the other variable. Another name for
this is a direct correlation.
1. What is the SAS syntax proc reg... used to evaluate, and what statistics do you

get from running this test that help you make a conclusion about your data?
a. Prog reg evaluates linear regression. After running the test, you get results that
include the y-intercept, slope, and r value to help you make a conclusion about your
data.
2

2. Which is stronger, a correlation of -0.85 or +0.55. Explain your answer.


a. Correlation of -0.85 is stronger because when comparing a positive correlation to a
negative correlation, you only look at the absolute numerical value. The higher the
numerical value (the closer to 1), the stronger the correlation. There is no need to
consider the positive or negative correlation when comparing the strength.

Hayley Hamilton and Rabab Zahidi


Fundamentals of Biostatistics Team Presentation
April 12th, 2016

Level 2:
3. For his biology project, Adam selected the research question Does the oxygen
level in water stimulate plant growth? State the dependent and independent
variables and explain their relationship.
a. The oxygen level will be the independent variable and the plant growth will be the
dependent variable. We want to quantify how the change in the oxygen level (IV) relates
to change in plant growth (DV).
4. The scatterplot below represents the data, as the temperature increases, fewer
hot chocolate products are sold. What variable is placed on the x-axis, and the
y-axis? Explain the correlation of the scatterplot. What is another name of the
correlation?

.
a. The x-axis contains the independent variable which is temperature, and the y-axis
consist of the dependent variable which hot chocolate products sold. Temperature
increase and sales of hot chocolate products are negatively correlated. As the
temperature increases there is a decrease in the sales for hot chocolate products.
Negative correlation is also known as inversely correlated.

Hayley Hamilton and Rabab Zahidi


Fundamentals of Biostatistics Team Presentation
April 12th, 2016

Level 3:
5. a) Demonstrate three ways that x and y could be related in this scenario.
b) Does temporal order exist in this scenario? Explain why or why not.
c) Why cant we investigate this question in an experimental setting?

a. 1) Divorce rates influences the per capita consumption of margarine. 2) The per
capita consumption of margarine influences the divorce rate. 3) A third variable (perhaps
husbands suggesting their wives eat butter instead) influences both variables.
b. No, temporal order does not exist in this scenario. We cannot know which came first
the change in divorce rate or the change in consumption of margarine.
c. It would be unethical to ask couples to get divorced for the sake of research.

Hayley Hamilton and Rabab Zahidi


Fundamentals of Biostatistics Team Presentation
April 12th, 2016

6. We ran the Honda civic data set in SAS using mileage as the independent
variable and price as the dependent variable. We got this data:

a. Compose a linear regression equation and identify r of this data.


y= 14801-0.07x
r =0.3471
2

b. What was the Null Hypothesis, do we accept or reject it and why?


The null hypothesis tests for the influence of mileage on price.
H : p=0 no association between mileage and price
H : p0 there is an association between mileage and price.
We reject the null hypothesis because p<0.05.
Therefore, there is an association between mileage and price,
0

c. Interpret the information found from this data:


For each one-mile increase on a Honda Civic, price significantly decreases by $0.07.
(p<.001)

Hayley Hamilton and Rabab Zahidi


Fundamentals of Biostatistics Team Presentation
April 12th, 2016

References
15 Insane Things That Correlate With Each Other. (n.d.). Retrieved April 7, 2016 from
http://tylervigen.com/spurious-correlations
Negative Correlation Examples. (n.d.). Retrieved April 7, 2016 from
http://examples.yourdictionary.com/negative-correlationexamples.html#ufTqWavjBQSTzccf.99
Questions the Linear Regression Answers. (n.d.). Retrieved April 7, 2016 from
http://www.statisticssolutions.com/questions-the-linear-regression-answers/

You might also like