Section 6 (Pgs 68-80) Flashcards
If the probability off successes at each binomial trial is p and number of trials is n, what does the mean number of successes equal?
np
If the probability of successes at each binomial trial is p and number of trials is n, what does the variance equal?
np(1-p)
If the probability of successes at each binomial trial is p and number of trials is n, what does the sample population proportion equal?
p
If the probability of successes at each binomial trial is p and number of trials is n, what does the sample standard error equal?
square root of p(1-p)/ n
Under what conditions can inferences about the number of successes, and the population proportion, be based on the normal distribution sampling theory?
when np and n(1-p) are both greater than 5
For a hypothesis test of binomial proportion, what 2 options can the test be based on?
The number of successes
The proportion of successes
What does regression analysis do?
Provides information about the nature of the relationship and enables predictions to be made
What does correlation technique do?
Assesses the linear relationship between 2 continuous variables
What technique is used to assess relationships between categorical variables?
Chi squared techniques
What technique is used to assess relationships when one variable is categorical?
Logistic regression
What is the word used to described when one set of continuous data changes proportionally with another?
Correlated
What measurement is used to describe the strength of the linear relationship between 2 continuous variables?
Correlation coefficient
What range of values can the correlation coefficient be between?
-1 and +1
What does the sign of the correlation coefficient indicate?
The direction of the relationship
What does the magnitude of the correlation coefficient indicate?
The strength of the relationship
If there is no linear association, what would the value of the correlation coefficient be?
0
Approximate value of the correlation coefficient if there is no correlation?
0-0.1
Approximate value of the correlation coefficient if there is a very weak positive correlation?
0.1-0.2
Approximate value of the correlation coefficient if there is a weak positive correlation?
0.2-0.4
Approximate value of the correlation coefficient if there is a moderate positive correlation?
0.4-0.7
Approximate value of the correlation coefficient if there is a strong positive correlation?
0.7-1.0
How would you describe it when there is a close relationship between the 2 variables with a large value of one variable being associated with a large value of the other?
Strong positive linear correlation
How would you describe it when the 2 variables cluster less closely around the line with a large value of one variable being associated with a large value of the other?
Moderate positive linear correlation
How would you describe it when there is a close relationship between the 2 variables with a large value of one variable being associated with a small value of the other?
Strong negative linear correlation
What is the full name of the correlation coefficient?
Pearson’s product moment correlation coefficient
What is the symbol for Pearson’s product moment correlation coefficient?
r
What is the initial steps in investigating relationships?
Produce a scatter plot
Does a mathematical relationship between 2 variables imply cause and effect?
No
Can correlation be used to measure agreement?
No
What happens to the value of r as sample size increases?
Value of r increases
What 3 conditions must be met for the significance of an observed value of r to be determined from tables?
At least one variable is plausibly normally distributed (preferably 2)
Measured on a random sample
Pairs of variables are independent
How can you tell if both variables on a scatter plot are normally distributed?
Scatter plot is elliptical
Does a mathematically significant correlation mean it is clinically significant?
No
What does r squared measure?
The proportion of the variation in the dependent variable (y) which is attributable to its linear relationship with the independent variable (x)
How is r squared dependent on the number of observations in the sample?
Independent of the number of observations
How should correlation be used to explore data?
Correlation should only be used to investigate specific hypothesis about relationships
Can correlation be used to measure agreement?
No - as agreement is comparing 2 values assessing the same quantity there is sure to be some sort of agreement
What can be plotted if there is a significant linear association between two variables?
The line of best fit
What is the general form of the regression line?
y = a + bx
What word can be used to describe a and b?
Constants
What is a?
The intercept
What is b?
The slope
What is the intercept?
The value of y when x=0
What is the slope?
The change in y when x increases by one unit
change in y/ change in x
What 3 assumptions are being made when regression methods are used?
The correlation between x and y variables is significant
For each value of the x variable, the values of the y variable have a normal distribution
The variances of these normal distributions are equal
Can the regression equation be used for values of the independent variable not increased in the sample range?
No
