Hypothesis Testing Assumptions

Nonparametric procedures may be applied when the data are measured on a ______ measurement scale

The test statistic in most cases is a ______ random variable

Nonparametric procedures have a reputation for requiring only ______ calculations when samples are large

The careful research worker always asks himself first whether he can use a ______ test

Because the assumptions are very ______, there is little problem of violation of assumptions

The process of collecting and compiling sample data may be ______ expensive and ______ time-consuming

Nonparametric tests, whenever they are applicable, are a great ______

The arithmetic can be ______ and ______ when samples are large and a computer is not available

According to the hypothesis testing, if the p-value is less than or equal to __________, we reject the null hypothesis.

The test statistic value calculated in step 3 is __________.

The Mann-Whitney U test is used for testing hypotheses about __________ populations.

One of the assumptions of the Mann-Whitney U test is that the variable observed is a __________ random variable.

The measurement scale should be at least __________ for the Mann-Whitney U test.

The conclusion of the hypothesis testing is that there is no sufficient evidence that the average number of hours is not equal to __________ hours.

If two or more |D𝑖 | are equal, assign each tied value the ______ of the rank positions occupied by the differences that are tied.

Assign the ______ of |D𝑖 | to its rank.

The test statistic is 1.______ or W- whichever is smaller (denoted by T) for H1: M  Mo.

The tabulated T for n and α/2 can be obtained using the ______ Signed Rank Table of Critical Values.

We will use the ______ method of hypothesis testing.

Reject Ho if ______ − value ≤ α, otherwise do not reject.

Test the hypothesis at the ______ level of significance that this particular lamp operates, on the average, 1.8 hours before requiring a recharge.

For the sake of an illustration, we will just use the ______ Signed Ranks test without testing the normality of the data or whether we met the assumptions of using t-test.

Error terms are not normally ______________.

Model fits all but one or few outlier ______________ are present.

Our objective in this subsection is to come up with a model that describes the relationship of the predictor variables with a continuous dependent variable in a ______________ fashion.

Not all variables hypothesized to affect the response variable may have a significant contribution in predicting the mean value of the response ______________.

For any set of k predictors, there will be a total of 2𝑘 possible number of models that can be ______________.

This calculation is based on the fact that each predictor can be either included or excluded from the ______________.

Among the competing models, a final model will be ______________.

It starts with all predictor variables in the ______________.

Correlated independent variables make it difficult to make inferences about the individual ______ coefficients (slopes) and their individual effects on the dependent variable (Y).

An independent variable known to be an important predictor ends up having a ______ regression coefficient that is not significant.

A regression coefficient that should have a ______ sign turns out to be negative, or vice versa.

When an independent variable is added or removed, there is a ______ change in the values of the remaining regression coefficients.

If two independent variables to be included in the model have a statistically significant ______ correlation, they are likely to cause multicollinearity problems.

If condition numbers are between 100 and 1000, there is ______ to strong collinearity.

A VIF > ______ indicates a multicollinearity problem.

One regressor variable is regressed on the remaining regressor variables and the ______ of multiple determination (R2) is computed.