Difference between revisions of "Manuals/calci/Bartlett'sTest"

From ZCubes Wiki
Jump to navigation Jump to search
(Created page with "<div style="font-size:25px">'''BARTLETTSTEST(DataRange,ConfidenceLevel,NewTableFlag)'''</div><br/> *<math>DataRange</math> is the array of x values. *<math>ConfidenceLevel</ma...")
 
Line 6: Line 6:
 
==Description==
 
==Description==
 
* Bartlett's test is used to test if k samples are from populations with equal variances.
 
* Bartlett's test is used to test if k samples are from populations with equal variances.
 +
* Also called Bartlett's test for homogeneity of variances. It is used to test if variances are equal for all samples.
 
* Bartlett's test is sensitive to departures from normality.  
 
* Bartlett's test is sensitive to departures from normality.  
 +
* It is used when you’re fairly certain your data comes from a normal distribution.
 
* That is, if the samples come from non-normal distributions, then Bartlett's test may simply be testing for non-normality.
 
* That is, if the samples come from non-normal distributions, then Bartlett's test may simply be testing for non-normality.
 
   <math>B=\frac{df_WlnMS_W-\sum_{j}df_jln s_j^2}{1+\frac{1}{3(k-1)}(\sum_{j}\frac{1}{df_j}-\frac{1}{df_W})}</math>
 
   <math>B=\frac{df_WlnMS_W-\sum_{j}df_jln s_j^2}{1+\frac{1}{3(k-1)}(\sum_{j}\frac{1}{df_j}-\frac{1}{df_W})}</math>

Revision as of 04:43, 8 May 2018

BARTLETTSTEST(DataRange,ConfidenceLevel,NewTableFlag)


  • is the array of x values.
  • is the value from 0 to 1.
  • is either TRUE or FALSE. TRUE for getting results in a new cube. FALSE will display results in the same cube.

Description

  • Bartlett's test is used to test if k samples are from populations with equal variances.
  • Also called Bartlett's test for homogeneity of variances. It is used to test if variances are equal for all samples.
  • Bartlett's test is sensitive to departures from normality.
  • It is used when you’re fairly certain your data comes from a normal distribution.
  • That is, if the samples come from non-normal distributions, then Bartlett's test may simply be testing for non-normality.
 
  • B is the Bartlett's test static.
  • is the pooled variance across all groups.

Result

  • If p-value is greater than BCriticl value, reject the null hypothesis.
  • Else retain null hypothesis.

Example

Spreadsheet
A B C D
1 51 82 79 85
2 87 91 84 80
3 50 92 74 65
4 48 80 98 71
5 79 52 63 67
6 61 79 83 51
7 53 73 85 63
8 54 74 58 93

=BARTLETTSTEST([A1:A8,B1:B8,C1:C8,D1:D8],0.05,true)

BARTLETT'S TEST
DATA-0 DATA-1 DATA-2 DATA-3
MEAN 60.375 77.875 78 71.875
VARIANCE 214.26785714285714 157.55357142857142 164.57142857142858 181.55357142857142
LNVARIANCE 5.367226901229239 5.059765536486956 5.1033446922005234 5.201550769540011
COUNT 8 8 8 8
DF 7 7 7 7
1/DF 0.14285714285714285 0.14285714285714285 0.14285714285714285 0.14285714285714285
ERROR
SAMPLE DATA
DF 28
1/DF 0.03571428571428571
VARIANCE 179.48660714285714
LNVARIANCE 5.19010059312721
RESULTS
DATA
B-NUMERATOR 0.19960131136474502
B-DENOMINATOR 1.0595238095238095
B 0.18838775454650092
P-VALUE 0.979441777737987
B-CRITICAL 7.810299999999978
RESULT THE P-VALUE IS LESSER THAN THE B-CRITICAL VALUE, SO THE VARIANCES ARE JUDGED TO BE EQUAL.