Photo AI

Last Updated Sep 27, 2025

Goodness of Fit Simplified Revision Notes

Revision notes with simplified explanations to understand Goodness of Fit quickly and effectively.

user avatar
user avatar
user avatar
user avatar
user avatar

457+ students studying

21.2.1 Goodness of Fit

χ2\chi^2 Goodness of Fit Tests

The χ2\chi^2 test can be used to test how well a distribution fits a data set.

lightbulbExample

Example: Eggs are sold in four categories: small, medium, large, and extra large. A supermarket model predicts that these will be sold in the ratio 1:2:3:1. To check this model, the supermarket looks at sales in a store in one day.

Size of eggsSmallMediumLargeExtra large
Number sold16172413

Use an appropriate statistical test to determine if the model fits this data, using a 5% significance level.

1. State hypotheses

  • H0H_0 : Eggs are sold in the ratio 1:2:3:11:2:3:1.
  • H1H_1 : Eggs are not sold in the ratio 1:2:3:11:2:3:1.

2. Calculate expectations by splitting total into the given ratio

  • Total =16+17+24+13=:highlight[70]= 16 + 17 + 24 + 13 = :highlight[70]
  • 7070 in the ratio 1:2:3:1:1:2:3:1:
  • 70÷7=1070 ÷ 7 = 10ratio=:highlight[10:20:30:10]ratio = :highlight[10:20:30:10] | Size of eggs | Small | Medium | Large | Extra large | |---|---|---|---|---| | Number sold | 16 | 17 | 24 | 13 | | Expected | 10 | 20 | 30 | 10 |

3. Perform a χ2\chi^2 test on the differences. Note that for goodness of fit tests, even when ν=1\nu = 1 , a Yates correction is never required.

Contributions:

(1610)210=3.6,(1720)220=0.45,(2430)230=1.2,(1310)210=0.9\frac{(16-10)^2}{10} = 3.6, \quad \frac{(17-20)^2}{20} = 0.45, \quad \frac{(24-30)^2}{30} = 1.2, \quad \frac{(13-10)^2}{10} = 0.9
  • See calculator screenshot for instructions.

Goodness of Fit on Graphical Calculator

Steps:

A) Go to Tests on the calculator and choose χ² GOF (Goodness of Fit).

B) Input observed and expected data in the respective lists (List 1, List 2).

C) After inputting the data, choose χ² GOF test.

D) Select List1 for observed values and List2 for expected values.

E) Input degrees of freedom (v=n1v = n - 1).

In this case, v = 3 because there are four categories (n=4n = 4), and the total has only one constraint, so degrees of freedom v=n1v = n - 1.

F) Execute the test by pressing EXE.

G) The calculator will display χcalc2=:highlight[6.15]\chi^2_{\text{calc}} = :highlight[6.15] and p=:highlight[0.10453613]p = :highlight[0.10453613].

H) χcalc2\chi^2_{\text{calc}} and pvaluep-value are shown.

I) Pressing Exit twice provides detailed results, including contributions in List 3.


4. Conclusion:

Since χ²calc = 6.15 < 7.815, we do not reject H0H_0.

Insufficient evidence to suggest that the ratio of eggs sold differs from the ratio 1:2:3:1.

Testing Hypothesis of Fit for Any Distribution

It is possible to test whether any known model fits a set of data.

Note: The model fits the data, not the data fits the model.

infoNote

Past Paper Example

Q4, (Jan 2008, Q4a)

In Germany, towards the end of the nineteenth century, a study was undertaken into the distribution of the sexes in families of various sizes. The table shows some data about the number of girls in 500 families, each with 5 children. It is thought that the binomial distribution B(5, p) should model these data.

Number of girlsNumber of families
032
1110
2154
3125
463
516

i) Use this information to calculate an estimate for the mean number of girls per family of 5 children. Hence show that 0.45 can be taken as an estimate of p.

ii) Investigate at a 5% significance level whether the binomial model with p estimated as 0.45 fits the data. Comment on your findings and also on the extent to which the conditions for a binomial model are likely to be met. [12 marks]


Solution: i)

Σxf=32×0+110×1+154×2+125×3+63×4+16×5=:highlight[1125]Σf=500soxˉ=1125500=:highlight[2.25]\Sigma xf = 32 \times 0 + 110 \times 1 + 154 \times 2 + 125 \times 3 + 63 \times 4 + 16 \times 5 = :highlight[1125]\\ \Sigma f = 500 \quad \text{so} \quad \bar{x} = \frac{1125}{500} = :highlight[2.25]μ=np2.25=5p(Since n=5 in this binomial model)p=2.255=:success[0.45](as required)\mu = np \quad \Rightarrow \quad 2.25 = 5p \quad \text {(Since n=5 in this binomial model)}\\ \Rightarrow \quad p = \frac{2.25}{5} = :success[0.45] \quad \text{(as required)}

Since we have estimated one of the population parameters, this means we have one less degree of freedom. Remember this point when checking critical values from the table.


Solution: ii)


Step 1: State hypotheses:

  • H0H_0: The proposed model fits the data well.
  • H1H_1: The proposed model does not fit the data well.

Step 2: Using the proposed model: Calculate the proportion of the total frequency associated with each outcome.

Using XB(5,0.45)X \sim B(5, 0.45):

P(X=0)P(X = 0) =0.05033= 0.05033

P(X=1)P(X = 1) =0.2059= 0.2059

P(X=2)P(X = 2) =0.3369= 0.3369

P(X=3)P(X = 3) =0.2957= 0.2957

P(X=4)P(X = 4) =0.1128= 0.1128

P(X=5)P(X = 5) =0.01845= 0.01845

Now, dividing the total frequency with the proportions, we get expectations:


Expectations:

| Number of girls | Number of families | | |---|---|---|---| | 0 | 25.165 | 500×0.05033\leftarrow \quad 500 \times 0.05033 | | 1 | 102.95 | 500×0.2059\leftarrow \quad 500 \times 0.2059 | | 2 | 168.45 | 500×0.3369, etc.\leftarrow \quad 500 \times 0.3369, \ etc. | | 3 | 137.85 | | | 4 | 56.4 | | | 5 | 9.225 | |

Notice no expectations < 5 \therefore no combining of items.


Observations:

Number of girlsNumber of families
032
1110
2154
3125
463
516

Step 3: Calculate χcalc2\chi^2_{\text{calc}} where contributions are calculated by:

Contributions=(OE)2E\text{Contributions} = \frac{(O - E)^2}{E}χcalc2=(3225.165)225.165+(110102.95)2102.95+(154168.45)2168.45+(125137.85)2137.85+(6356.4)256.4+(169.225)29.225\chi^2_{\text{calc}} = \frac{(32 - 25.165)^2}{25.165} + \frac{(110 - 102.95)^2}{102.95} + \frac{(154 - 168.45)^2}{168.45}+\frac{(125 - 137.85)^2}{137.85} + \frac{(63 - 56.4)^2}{56.4} + \frac{(16 - 9.225)^2}{9.225}

Note: If asked to analyze contributions, it is necessary to calculate the value of each individual contribution.

=1.856+0.4828+1.240+1.198+0.7723+4.976= 1.856 + 0.4828 + 1.240 + 1.198 + 0.7723 + 4.976:highlight[10.525]\approx :highlight[10.525]

Step 4: Check the critical value and conclude appropriately.

  • Remember:
v=(no. of items after combining)(no. of estimated parameters)1v = (\text{no. of items after combining}) - (\text{no. of estimated parameters}) - 1

v=611=:highlight[4](because we estimated p)v = 6 - 1 - 1 = :highlight[4] \quad \text{(because we estimated } p)

Critical Value (C.V.) from the table:

C.V.=:highlight[9.488]\text{C.V.} = :highlight[9.488]

Calculated value:

:highlight[10.525>9.488]:highlight[10.525 > 9.488]

Conclusion: Reject H₀.

The binomial model is not a good fit for the data.

In the proposed model, we seem to underestimate in the extremes and overestimate in the middle.

The biggest contribution is for X=5X = 5 , indicating that this model is a poor fit, especially at the right-hand tail.

Within a family, the sex of one child may not be statistically independent of a previously born child. Also, the probability of giving birth to a girl is unlikely to be 0.450.45 across all families. Therefore, the binomial model may not be appropriate.

Books

Only available for registered users.

Sign up now to view the full note, or log in if you already have an account!

500K+ Students Use These Powerful Tools to Master Goodness of Fit

Enhance your understanding with flashcards, quizzes, and exams—designed to help you grasp key concepts, reinforce learning, and master any topic with confidence!

40 flashcards

Flashcards on Goodness of Fit

Revise key concepts with interactive flashcards.

Try Further Maths Further Statistics 1 Flashcards

4 quizzes

Quizzes on Goodness of Fit

Test your knowledge with fun and engaging quizzes.

Try Further Maths Further Statistics 1 Quizzes

29 questions

Exam questions on Goodness of Fit

Boost your confidence with real exam questions.

Try Further Maths Further Statistics 1 Questions

27 exams created

Exam Builder on Goodness of Fit

Create custom exams across topics for better practice!

Try Further Maths Further Statistics 1 exam builder

50 papers

Past Papers on Goodness of Fit

Practice past papers to reinforce exam experience.

Try Further Maths Further Statistics 1 Past Papers

Other Revision Notes related to Goodness of Fit you should explore

Discover More Revision Notes Related to Goodness of Fit to Deepen Your Understanding and Improve Your Mastery

96%

114 rated

Chi Squared Tests

Product Moment Correlation Coefficient

user avatar
user avatar
user avatar
user avatar
user avatar

229+ studying

195KViews

96%

114 rated

Chi Squared Tests

Chi Squared Tests for Standard Distributions

user avatar
user avatar
user avatar
user avatar
user avatar

301+ studying

200KViews

96%

114 rated

Chi Squared Tests

Chi Squared Tests for Contingency Tables

user avatar
user avatar
user avatar
user avatar
user avatar

475+ studying

182KViews
Load more notes

Join 500,000+ A-Level students using SimpleStudy...

Join Thousands of A-Level Students Using SimpleStudy to Learn Smarter, Stay Organized, and Boost Their Grades with Confidence!

97% of Students

Report Improved Results

98% of Students

Recommend to friends

500,000+

Students Supported

50 Million+

Questions answered