Chi Square

Question 1

<p>Chi square is sometimes used to determine what in regards to two variables??</p>

Answer 1

<p>Whether or not two variables are independent of each other or if they are related to each other.</p>

Question 2

<p>What is a contingency table?</p>

Answer 2

<p>A two way table showing the contingency between two variables where the variables have been classified into mutually exclusive categories and the cell entries are frequencies.</p>

Question 3

<p>Define contingency!</p>

Answer 3

<p>An event that may, but is not certain to occur aka something that may happen.</p>

Question 4

<p>What is a parametric test?</p>

Answer 4

<p>One that depends considerably on population characteristics or parameters for its use. (mean, SD etc)</p>

Question 5

<p>If nonparametric tests like Chi Square do not carrymany assumptions and characteristics then why not use them all of the time.</p>

Answer 5

<p>Parametric tests are much more powerful and more versatile than nonparametric tests like Chi Square because they use more information in the calculation (more than just frequency).</p>

Question 6

<p>What is the statistical inference used most often with nominal data?<br></br>Hint: It is the only test we will cover in this class for nominal data.</p>

Answer 6

<p>The nonparametric test = chi square (X^2)</p>

Question 7

<p>What are the steps for calculating Chi Square?</p>

Answer 7

<p>Take the sum of the square of each observed frequency less the expectived frequency and divide by the expected frequency.</p>

<p>x^{2 =}<strong>∑ </strong>(Oi - Ei)²/ Ei</p>

<p></p>

Question 8

<p>What are the two types of statistical tests?</p>

Answer 8

<p>1. Parametric 2. Nonparametric </p>

Question 9

<p>What are the two assumptions of a parametric test?</p>

Answer 9

<p>1. The scores of the population are normally distributed</p>

<p>2. The variance across groups is homogenous.</p>

Question 10

<p>What level of data is used for<em>X</em>^2?</p>

Answer 10

<p>When dealing with nominal (categorical data).</p>

Question 11

<p>What are the three assumptions underlying Chi Square?</p>

Answer 11

<p>1. There is independence between each observation recorded in the contingency table. A) each subject may have only one entry in the table</p>

<p>2. Sample size must be large enough that the expected frequency in each cell is at least five for tables where there are more than two rows or columns</p>

<p>3. Normally used with nominal data A) however, can be used with ordinal, interval or ratio data.</p>

Question 12

<p>Is the following hypothesis a research or null hypothesis?</p>

<p>There is no difference in satisfied verses dissatisfied couples with respect to the number of unresolved arguments.</p>

Answer 12

<p>This is a null hypothesis or Ho, A null hypothesis indicates that there is no difference.</p>

Question 13

<p>Describe what the frequency expected ( fe) is ??</p>

Answer 13

<p>Expected frequency under the assumption sampling is random from the population. It is what we expect to get in each cell.</p>

Question 14

<p>Describe what the frequency observed is! (fo)?</p>

Answer 14

<p>This is the observed frequency for a particular cell.</p>

Question 15

<p>X^2 allows analysis of the difference scores between \_\_\_ and \_\_\_\_.</p>

Answer 15

<p>frequency expected and frequency observed.</p>

Question 16

<p>True or False<br></br>You can only use the Chi Square Test for nominal data?</p>

Answer 16

<p>False</p>

<p>You can treat ordinal or interval data as categorical and perform a Chi Square Test.</p>

Question 17

<p>When is a Chi Square Test useful in terms of NOIR?</p>

Answer 17

<p>1. For nominal data.</p>

<p>2. For higher levels of data when the distribution is skewed or kurtotic (positve skew, negative skew, leptokurtic, platykurtic).</p>

<p>3. If the group sizes are unequal. (Some parametric tests are sensitive to sample size).</p>