An Introduction to Statistics - Lesson 10

Chi Square Goodness of Fit

Lesson Overview

• Chi Square (²) Distribution and Tests
• Table of ²) Values
• Other Applications

Chi Square Distributions and Tests

The ² family of distributions is characterized by one parameter called the degrees of freedom which is often denoted by v (the greek letter nu) and used as a subscript: ²_v.

1. The ² distribution is continuous.
2. The ² distribution is unimodal.
3. The ² distribution is always positive (> 0).
4. The ² distribution mean = v.
5. The ² distribution variance = 2v.
6. For small v (v < 10), the distribution is highly skewed to the right (positive).
7. As v increases the ² distribution becomes more symmetrical about v (the mean).
8. We can thus approximate the ²_v when v > 30 with the normal (see table below).

Gosset first described the distribution of s². It is related to the ² by the simple factor (n-1)/². Although he wasn't able to prove this mathematically, he demonstrated it by dividing a prison population of 3000 into 750 random samples of size four and used their heights.

A common application of the ² distribution is in the comparison of expected with observed frequencies. When there is but one nominal variable, this is often termed goodness of fit. In this case we are testing whether or not the observed frequencies are within statistical fluctuations of the expected frequencies. Although one typically checks for high ² values, the second example below illustrates the possible significance of a low ² value.

Example: On July 14, 2005 the EDRM611 class collected 10 trials of 20 pennies each where these 20 pennies were set on edge and the table banged. The class observed 145 heads. We can compare the observed with expected frequencies and test for goodness of fit as shown in the table below. There is but one degree of freedom since the number of tails is dependent on the number of heads (200 - 145 = 55).

Solution: We form the ² statistic by summing the (O-E)²/E and get 2045/100 + 2045/100 = 40.9. We can then compare this ² with critical ² values or find an associated P-value. The critical ² value for df=1 and one-tailed, alpha=0.05 is 3.841. Our results are far to the right of 3.841 so are VERY significant (P-value=1.6×10^-10). A table of critical ² values for select values is given below.

A Chi Square Distribution Table

Example: On July 12, 2005 the EDRM611 class collected 192 dice rolls, each person present using a different die and each person doing 24 rolls. Were the results within the expected range?

Solution: We form the ² statistic by summing the (O-E)²/E and get 208/32=6.5. We can then compare this ² with a critical ². Only if it is more extreme is it worth finding a P-value. We have 6 - 1 = 5 degrees of freedom. The critical ² values for df=5, two-tailed, and alpha=0.05 are 1.145 and 11.07. Since our ² is within this range, our results are within the range we can expect to occur by chance. Notice the lower ² cut off. When people fabricate a random distribution they are likely to make it too uniform and get too small of a ² which can be checked as above, but the ² would likely be less than 1.145. Working backwards we see the sum of the (O-E)² would have to be less than 36 so if one were 5 or less away and the rest much closer, we might wonder.

As noted at the bottom of the table above, when the degrees of freedom are large, a z-score can be formed and compared against a standard normal distribution. Note also that the mean of any ² is the degrees of freedom. This might be helpful to realize where the distribution is centered.

The ² goodness of fit does not indicate what specifically is signficant. To find that out one must calculate the standardized residuals. The standardized residual is the signed square root of each category's contribution to the ² or R = (O - E)/sqrt(E). When a standardized residual has a magnitude greater than 2.00, the corresponding category is considered a major contributor to the significance. (It might be just as easy to see which (O - E)²/E entries are larger than 4, but standardized residuals are typically provided by software packages.)

Other Applications

There are potential problems associated with small expected frequencies in contingency tables. Historically, when any cell of a 2×2 table was less than 5 a Yates' correction of continuity was advised. However, it has been shown that this can result in a loss of power (a tendancy not to reject a false null hypothesis). Care should be exercised and advise sought. Larger contingency tables can also be problematic when more than 20% of the cells have expected frequencies less than 5 or if there are any cells with 0. One solution is to combine adjacent rows or columns, but only if it makes sense.

In closing we should note the importance of focusing on a small number of well-conceived hypotheses in research rather than blindly calculating a bevy of ² statistics for all variable pairs and ending up with 5% of your results being significant at the 0.05 level! You would even expect 1% of your results, due to pure random chance in your sample selection, to be significant at the 0.01 level. Since there are n(n - 1)/2 possible pairings for n variables, one would have 5050 pairs for 100 variables of which over 250 could look significant at the 0.05 level. Beware!

This concludes our introduction to statistics. We will continue next year with combinatorics, a survey of distributions, and more inferential statistics. Please collect your lectures and homework for stapling.

BACK	HOMEWORK	ACTIVITY	CONTINUE

• e-mail: calkins@andrews.edu
• voice/mail: 269 471-6629/ BCM&S Smith Hall 106; Andrews University; Berrien Springs,
• classroom: 269 471-6646; Smith Hall 100/FAX: 269 471-3713; MI, 49104-0140
• home: 269 473-2572; 610 N. Main St.; Berrien Springs, MI 49103-1013
• URL: http://www.andrews.edu/~calkins/math/webtexts/stat10.htm
• Copyright ©1998–2005, Keith G. Calkins. Revised on or after October 9, 2005.