Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
118 Cards in this Set
- Front
- Back
|
Does testing the null hypothesis also called significance testing?
|
No.
|
|
|
What do we mean when a sample mean or sample percentage is significant?
|
Any mean or percentage above 1.96 or below -1.96 is significant.
|
|
|
What is hypothesis testing?
|
Hypothesis testing is testing a significance of a mean or percentage.
|
|
|
What are the two possible outcomes when testing the null hypothesis?
|
Either we reject the null hypothesis or we fail to reject the null hypothesis.
|
|
|
When do we reject the null hypothesis?
|
When the sample falls OUTSIDE of the 95% confidence interval.
|
|
|
What is a=.05
|
a= alpha. .05 means that we set an assumption that our sample mean or percentage is significantly different from the mean or percentage of the null hypothesis if it falls in the 5% of the samples that have means or percentages that are furthest away from the hypothesized population mean or percentage.
|
|
|
If the alpha level is .05, then the critical z or t is what?
|
Is the actual standard errors that seperates sample means or percentages that are 5% away from other sample means or percentages that are closer to the mean or percentage as suggested by the null hypothesis.
|
|
|
In a significance test presented from this powerpoint presentation, do we reject the z or t if it is above 1.96 or below -1.96?
|
Yes
|
|
|
Is there a difference between t and z when we work with sample sizes above 120 people?
|
No. The difference between t and z does not matter unless we are working with samples that are LESS than 120 people.
|
|
|
What are critical t values?
|
-1.96 and 1.96 (remember that t is basically z when N is OVER 120).
The critical t values seperates the 5% of the sample means or percentages that would be furthest away from the hypothesized population mean or percentage if we theoretically did repeated sampling. The t values are used to reject or fail to reject the null hypothesis. |
|
|
What are at least two reasons that critical chi squared (x2), values are different from critical t values?
|
1. We build a 95% confidence interval surrounding the null hypothesis of x2=0. Yet, there is only ONE critical value: x2 is chi squared so you can't have a negative critical value. To reject the null hypothesis, your sample x2 has to be ABOVE the critical x2.
2. The critical x2 always changes,d epending on the number of categories of each nominal/ordinal variable of a bivariate relationship. You will learn to determine the critical x2 to test the null hypothesis by considering the number of categories for each nominal/ordinal variable in a bivariate relationship. |
|
|
What can be the levels of measurement for the dependent variables for cross tabs?
|
nominal, ordinal, dichotomous.
|
|
|
What can be the levels of measurement for independent variables for cross tabs?
|
nominal, ordinal, dichotomous.
|
|
|
In cross tabs, are we asking whether there is a relationship between two nominal or ordinal variables?
|
Yes, and this includes dichotomous variables.
|
|
|
Are means and standard deviations meaningless for cross tabs tests?
|
Yes.
|
|
|
What does a chi-squared, x2, tell us?
|
Tells us whether the numbers in our sample deviate from the null hypothesis that there is no relationship between variables.
|
|
|
What does a big chi-square tell us about the relationship between two nominal variables?
|
Will imply that there IS a relationship between our two nominal variables.
|
|
|
If the chi-squared statistic is significant, are your two nominal variables related?
|
Yes.
|
|
|
What is the value of the slope of a flat, horizontal line?
|
zero.
|
|
|
What can be the levels of measurements for the dependent variable for correlation and regression?
|
Interval-ratio
|
|
|
What can be the levels of measurements for the independent variable for correlation and regression?
|
Interval-ration and dichotomous.
|
|
|
What is a continuous variable?
|
A variable can have values of 1, 2, 3, 4 as well as in between such as 1.3, 2.7, 3.5, and so forth. Length, weight, time are examples of continuous variables.
|
|
|
What is a discrete variable?
|
A variable can only take on whole values such as 1,2,3,4 and so forth. An example is gender in which you code female as "1" and male as "0". Political party is another example. Democrat "1" republican "0".
|
|
|
What is a bivariate relationship?
|
Two variables are related to (associated with) each other.
|
|
|
What is a correlation?
|
A statistic that assesses the strength and direction of linear association of two interval/ratio variables...it is created through a technique called "regression"
|
|
|
What is bivariate regression?
|
A technique that fits a straight line as close as possible between all the coordinates of two interval/ratio variables plotted on a two dimensional graph-to summarize the relationship between the variables.
|
|
|
Does the slope of regression tell us whether relationship between two variables is positive, negative, or no relationship?
|
Yes.
|
|
|
What does OLS mean?
|
Ordinary Least Squares
|
|
|
Do we use squared deviations to fit a regression line that is closest to all data points?
|
Yes.
|
|
|
What is the range in a (Pearson) correlation?
|
-1 to 1.
|
|
|
What is the value of a correlation when there is a perfect positive correlation?
|
1
|
|
|
What is the value of a correlation when there is a perfect negative correlation?
|
-1
|
|
|
What is the value of a correlation when there is no relationship between two variables?
|
Zero.
|
|
|
What can R2(squared) be interpreted as?
|
IS called the "coefficient of determination".
The improvement obtained by using X (by drawing a line through the conditional means) in getting as near as possible to everybody's value for Y over just using the mean for Y alone. |
|
|
Can a slope be called a coeffeicient?
|
Yes.
|
|
|
In multiple regression, are there two or more interval-ratio or dichotomous variables as independent variables?
|
Yes
|
|
|
If we looking at the affects of education and family income on the number of children, what are the two null hypothesis?
|
1. There is no relationship between education of respondents and the number of children in families.
Ho: B1=0 2. There is no relationship between family income and the number of children in families Ho: b2=0 |
|
|
What are dummy variables?
|
They are simply dichotomous variables that are entered into regression. They have 0-1 coding where
0= absence of something 1=presence of something Ex: Female (0=M, 1=F) or Southern (0=Non-Southern 1=Southern). |
|
|
What are two values for dummy variables?
|
0 and 1.
|
|
|
Do you leave out one of the dummy variables in a multiple regression equation?
|
Yes.
|
|
|
In the regression equation in which the dependent variable is self-esteem, which race variable is left out?
|
White.
|
|
|
In the regression equation in which the dependent variable is self-esteem how many points are lower for women's self-esteem compared to men's self-esteem?
|
4 points.
|
|
|
In the regression equation in which the dependent variable is self-esteem, how many points are lower for Blacks self-esteem compared to white's self-esteem?
|
Black's self-esteem is 5 points HIGHER than whites.
|
|
|
In the regression equation in which the dependent variable is self-esteem, how many points are lower for other's self-esteem compared to white's self-esteem?
|
Other's self-esteem is 2 points LOWER than white's and consequently 7 points lower than blacks.
|
|
|
In the regression equation in which the dependent variable is self-esteem, each year of education improves self-esteem by how many units?
|
0.3 units
|
|
|
What is the self-esteem score of white males with 10 years of education?
|
33
|
|
|
What is the self-esteem score of black males with 10 years of education?
|
38
|
|
|
What is the self-esteem score of white females with 20 years of education?
|
28.92
|
|
|
What is ethics?
|
Rules of proper and improper conduct of research.
|
|
|
What is involuntary participation?
|
When participants are coerced into doing research.
|
|
|
What is informed consent?
|
An understanding by research participants f the possible risks involved participating in social research.
|
|
|
What is anonymity?
|
When the researcher or readers of research do not know the identification of the research participants.
|
|
|
What is confidentiality?
|
When the researcher promises not to reveal the research participants identities to readers of research.
|
|
|
What is deception?
|
Lying to research participants to test research questions. It is often set in a laboratory setting so that the research participant knows that some type of research is being conducted.
|
|
|
What is debriefing?
|
An option used after deception is conducted. Means that the researchers interviews that participant about his or her research experience. The researcher can reveal the true purpose of the research to the participant. The research would discover and remedy any bad feelings experienced by the participant.
|
|
|
Does the political and social climate influence the type of research being conducted?
|
Yes.
|
|
|
What is the range of the Pearson r?
|
-1 to 1.
|
|
|
Does -1.00 indicate the perfect inverse relationship?
|
Yes.
|
|
|
Does 1.00 indicated the perfect direct relationship?
|
Yes.
|
|
|
Does 0.00 indicate the absence of a relationship?
|
Yes
|
|
|
The closer a value to 0.00, is the relationship stronger?
|
no.
|
|
|
The closer a value to -1.00 or 1.00, is the relationship stronger?
|
Yes.
|
|
|
Is r a proportion?
|
Yes.
|
|
|
Is it possible for a relationship to be both direct and weak?
|
No.
|
|
|
Is it possible for a relationship to be both inverse and strong?
|
No.
|
|
|
Which of the following values of r represents the strongest relationship?
.64, -.79, or 0.00? |
-.79
|
|
|
How do we obtain the coefficient of determination?
|
prediction of future outcomes on the basis of other related information. It is the proportion of variability in a data set that is accounted for by the statistical model.It provides a measure of how well future outcomes are likely to be predicted by the model.
|
|
|
What is the coefficient of determination when converted to a percentage?
|
67%
|
|
|
When r=40, is the percentage of variance accounted for 16%? (Remember to square and convert to percentage).
|
Yes.
|
|
|
When r=.40, is the percentage of variance NOT accounted for 84%? (Remember to square, convert to percentage, and subtract from 100%)
|
Yes.
|
|
|
When r=.80, what is the percentage of variance accounted for?
|
64%
|
|
|
When r=.80, what is the percentage of variance NOT accounted for?
|
36%
|
|
|
What does a scattergram illustrate?
|
It gives a good visual picture of the relationship between the two variables, and aids the interpretation of the correlation coefficient or regression model.
|
|
|
When the dots on the scattergram are more scattered, is the relationship between the two variables weaker?
|
Yes.
|
|
|
If the dots in a scattergram forma pattern from the lower left to the upper right, is the relationship inverse?
|
No, direct.
|
|
|
What are random (chance) errors known as?
|
Simple random sampling
|
|
|
What does the central limit theorem say?
|
states conditions under which the mean of a sufficiently large number of independent random variables, each with finite mean and variance, will be approximately normally distributed
|
|
|
What term is used to describe the standard deviation of a sampling distribution?
|
standard error
|
|
|
If the sample is larger, the standard error of the mean is smaller or larger?
|
Smaller
|
|
|
If the sample is smaller, the standard error of the mean is smaller or larger?
|
Larger
|
|
|
What percentages are more commonly reported on confidence intervals?
|
95%, 99%.
|
|
|
Does using a large sample keep the standard error of a mean and the confidence intervals small?
|
Yes.
|
|
|
Can the difference between two means be due to sampling error?
|
Yes.
|
|
|
Is this an appropriate null hypothesis: "thee is not true difference between the means"
|
Yes.
|
|
|
What is the symbol for the null hypothesis?
|
H0
|
|
|
What type of hypothesis predicts that one particular's group's mean will be higher than the other group's mean?
|
Null Hypothesis.
|
|
|
What does a significance test yield?
|
Whether our sample represents a specific population as suggested by a null hypothesis, using our sample mean or percentage.
|
|
|
If the p is less than 5 in 100 in a null hypothesis, do we conclude that null hypothesis is not likely to be true?
|
Yes
|
|
|
What is a type 1 error?
|
Rejecting the null hypothesis.
|
|
|
What is a type 2 error?
|
innapropriately retaining the hypothesis.
|
|
|
What is the synonym for rejecting the null hypothesis?
|
type 1 error.
|
|
|
Is the .05 level or the .01 level more significant?
|
.01
|
|
|
When p>.05, is the difference between two means usually regarded as "statistically significant" or "statistically insignificant?"
|
Statistically insignificant
|
|
|
When the t test yields a low probability that the null hypothesis is correct, do researchers usually reject the null hypothesis?
|
yes.
|
|
|
The larger the sample, is the null hypothesis more likely to be rejected?
|
yes
|
|
|
The larger the observed difference between the two means, is the null hypothesis is more likely to be rejected?
|
yes.
|
|
|
The smaller the variance, is the null hypothesis likely to be rejected?
|
yes.
|
|
|
Is the null hypothesis rejected when p>.05?
|
No.
|
|
|
What does ANOVA stand for?
|
Analysis of variance between groups.
|
|
|
What is ANOVA used for?
|
provides a statistical test of whether or not the means of several groups are all equal, and therefore generalizes Student's two-sample t-test to more than two groups. ANOVAs are helpful because they possess a certain advantage over a two-sample t-test. Doing multiple two-sample t-tests would result in a largely increased chance of committing a type I error. For this reason, ANOVAs are useful in comparing three or more means.
|
|
|
What test is ANOVA used for?
|
t-tests. (two sample).
|
|
|
Do we reject the null hypothesis for ANOVA when one or more differences in means is statistically significant?
|
yes.
|
|
|
What is a direct relationship?
|
elationship between two variables in which change in one variable is associated with a change in the other variable in the same direction.
|
|
|
Is a direct relationship the same thing as a positive relationship?
|
Yes.
|
|
|
What is an inverse relationship?
|
in which one variable, say y, decreases as another, say x, increases.
|
|
|
Is an inverse relationship the same thing as a negative relationship?
|
Yes.
|
|
|
For the scores on test X and test Y shown below, the correlation is..?
a. direct b. inverse c. no relationship |
B. Inverse relationship
|
|
|
What is the value of the null hypothesis between two variables?
|
Zero
|
|
|
In ANOVA, can you compare the means between 3 or more groups?
|
Yes.
|
|
|
What is BSS?
|
Between Sum of Squares
|
|
|
What is WSS?
|
Within Sum of Squares
|
|
|
In ANOVA, what is the null hypothesis?
|
The sample comes from a population where there are no significant differences between the degree groups on the mean age of the respondent when his or her first child is born.
|
|
|
What is the standard null hypothesis between two variables?
|
No relationship between two variables.
|
|
|
Do we reject the null hypothesis when the t-statistic is ABOVE 1.96 and -1.96?
|
Yes
|
|
|
Do we reject the null hypothesis when the p value is less than .05?
|
Yes
|
|
|
What is the level of measurement for the independent variable for an independent samples t test?
|
Dichotomous
|
|
|
What is the level of measurement for the dependent variable for an independent samples t test?
|
interval, ratio, dichotomous.
|
|
|
What is an independent samples t test concerned with?
|
Whether a mean or proportion is equal between two groups.
|
