Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Xbar Chart: Farrokh Alemi, PH.D

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 13

Xbar Chart

Farrokh Alemi, Ph.D.


Why Chart Data?
 To discipline intuitions
 To tell a story
Steps to create an
X-bar chart?
1. Check assumptions
2. Calculate grand average
3. Calculate standard deviation
4. Calculate standard deviation for each time
period
5. Calculate control limits
6. Plot chart
7. Interpret and distribute
Example Data
 Over several
months, we
tracked the Time
Ratings of Ratings of Ratings of Ratings of
1st 2nd 3rd 4th
Period
satisfaction with Subject Subject Subject Subject
1 80 84 82 80
our unit. 2 70 72 74 70
3 76 78 76 78
 The question is 4 80 78 78 80

whether the unit


has improved
over time.
Are variations due
to chance?
 How would you answer
85
this question? Look at

Satisfaction rating
80
the data.
75
 There are wide 70
variations in the data. 65
Could these variations 1 2 3 4
be due to chance? Time period

 The first step is to


calculate the average
for each time period.
1. Check
Assumptions
 Continuous Interval Scale
 Independent events

 Normal distribution

 Constant variance
2. Calculate Grand
Average
 Average values for
all time periods and
all cases, plot as a
line
85
Satisfaction rating

80
Observed
75
Grand average
70

65
1 2 3 4
Time period
3. Calculate
standard deviation
Calculate standard deviation for all
observations
S = [i=1,…,nj j = 1, …m (Xij -GA )2 / (n-1)]0.5
GA is grand average
n is number of observations across time
periods
4. Calculate
standard deviation
for each time period
Estimate the standard deviation for each
time period
St = S / (nt)0.5
5. Calculate Control
Limits
UCL = GA + 1.96 * St
LCL = GA – 1.96 * St

 1.96 sets the limits so that 95% data fall


within the two limits
 GA is the grand average
 St is the standard deviation for the time
period
Calculations in
Excel
Constant
for 95%
Grand confidence Standard deviation
average limits function over range of
all observations

=Average( =Average Dividing by square root


B2:E2) ($B$2:$E$5) of number of
observations in the
time period
6. Plot Control Chart
85
Satisfaction rating

80
Observed
75 UCL
LCL
70

65
1 2 3 4
Time period
7. Interpret findings
& distribute
 First time period is
above UCL 85

Satisfaction rating
 Second time period 80
Observed
75 UCL
below LCL 70
LCL

65
1 2 3 4
Time period

You might also like