Question

Question 1

For correlation, you will ask if the slope of the line relating variable X and variable Y is significantly different from zero

1. True
2. False

Question 2

For regression, you test if the R2 value is significantly different from zero

1. True
2. False

Question 3

For correlation, the data are assumed to be normally distributed in both the X and the Y directions.

1. True
2. False

Question 4

For regression, the data are assumed to be normally distributed in both the X and Y directions.

1. True
2. False

Question 5

For correlation, the alternative hypothesis can be left, right or both tailed depending on the question

1. True
2. False

Question 6

For regression, the alternative hypothesis can be left, right or both tailed depending on the question

1. True
2. False

Question 7

Nonparametric tests use the original data to analyze the question being asked

1. True
2. False

Question 8

In class we covered a nonparametric tests for every parametric test covered this semester

1. True
2. False

Question 9

Nonparametric tests still have the assumptions of a 'good' sample

1. True
2. False

Question 10

For the Spearman's Rank Test, there still has to be a pattern of covariance in ranks to find a significant relationship between your X and Y variables.

1. True
2. False

Answer 1

Answer1: TRUE

Reason: Actually Slope of a line measurment of how many unit's it goes up or down for every unit we move the right .

In general , straight lines have slope that are positive negative or zero. If X and Y are negatively related then slope is negative and if X and Y are  positively related  then slope is positive and if slope is zero then no relationship between X and Y.

Answer2: True

Reason: R^2 which denote the coefficient of dittermination that is its measure the how independent variable explained the dependent variable . If R^2 greater that zero it means it measure how much  % explained by independet variable (X)for dependent variable ( Y). Suppose R^2=.7 then its means X explained 70% of Y.

Answer3: False

Reason: For Correlation, when we estimate the parameter(estimation process) with least square method then we don't need assumption of normal of data for X and Y. But when we use MLE method then we need of normalty assumption .At a time of testing of hypothesis we need assumption of normalty.

Answer 4 : False

Reason: For Regressiom, when we estimate the parameter(estimation process) with least square method then we don't need assumption of normal of data for X and Y. But when we use MLE method then we need of normalty assumption .At a time of testing of hypothesis we need assumption of normalty.

Answer5: True

Reason: Null hypothesis for any hypothesis testing is simple hypothesis that is equality, But alternative hypothesis depends on different situation of question that what the asked. Its 3 type, one is Right tailed test , second is left tailed test and thirdone is mixed tailed test.

Answer6: True

Reason: Null hypothesis for any hypothesis testing is simple hypothesis that is equality, But alternative hypothesis depends on different situation of question that what the asked. Its 3 type, one is Right tailed test , second is left tailed test and thirdone is mixed tailed test.

Answer7: False

Reason: Non parametric test used of sampled data which comes from original data that is comes from population. sampled data should be independent and does'nt assume data comes from normal population.

Answer8: False

Reason: Non-parametric tests are valid for both non-Normally distributed data and Normally distributed data But due to accuracy , time and validness we use parametric test for ordinal or nominal data and parametric test for interval and ratio scale and also we take assumption of normalty for parametric test.

Answer9: False

Reason: Non parametric test used of sampled data which comes from original data that is comes from population. sampled data should be independent and does'nt assume data comes from normal population.

Answer10: True

Reason: As Karl Pearson's coefficient of correlation measure a linear relatioship between quantitative relationship between two variable , In similar way Spearman's coefficient of correlation between rank xi's and yi's, That is generally for qualitative data.