stats chap03 bluman ppt
TRANSCRIPT
-
8/16/2019 Stats Chap03 bluman ppt
1/86
Chapter 3
Data Description
1© McGraw-Hill, Bluman, 5
th
ed, Chapter3
-
8/16/2019 Stats Chap03 bluman ppt
2/86
Chapter 3 Overview
Introduction
3- Measures o! Central "endenc#
3-$ Measures o! %ariation
3-3 Measures o! &osition
3-' ()plorator# Data *nal#sis
2Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
3/86
Chapter 3 O+ectives
.ummari/e data usin0 measures o!central tendenc#
$ Descri+e data usin0 measures o!
variation3 Identi!# the position o! a data value in a
data set
' 1se +o)plots and !ive-num+ersummaries to discover various aspectso! data
3Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
4/86
Introduction
Traditional Statistics
Average
Variation
Position
4Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
5/86
3 Measures o! Central "endenc#
* statisticstatistic is a characteristic or measureo+tained +# usin0 the data values !rom asample
* parameter parameter is a characteristic ormeasure o+tained +# usin0 all the data
values !or a speci!ic population
5Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
6/86
-
8/16/2019 Stats Chap03 bluman ppt
7/86
-
8/16/2019 Stats Chap03 bluman ppt
8/86
Measures o! Central "endenc#4
Mean "he meanmean is the uotient o! the sum o!the values and the total num+er o! values
"he s#m+ol is used !or sample mean
6or a population, the Gree7 letter μ 8mu9is used !or the mean
X 1 2 3 n
X X X X X X
n n
+ + + += =
∑L
1 2 3 N X X X X X
N N µ
+ + + += =
∑L
8Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
9/86
Chapter 3
Data Description
.ection 3-()ample 3-
&a0e :;<
9Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
10/86
()ample 3-4 Da#s O!! per =ear
"he data represent the num+er o! da#s o!! per#ear !or a sample o! individuals selected !romnine di!!erent countries 6ind the mean
$;, $
-
8/16/2019 Stats Chap03 bluman ppt
11/86
Rounding Rule: Mean
"he mean should +e rounded to one moredecimal place than occurs in the raw data
"he mean, in most cases, is not an actualdata value
11Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
12/86
Measures o! Central "endenc#4
Mean !or Grouped Data "he mean !or 0rouped data is calculated
+# multipl#in0 the !reuencies and
midpoints o! the classes
m f X X
n
×=
∑
12Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
13/86
Chapter 3
Data Description
.ection 3-()ample 3-3
&a0e :;>
13Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
14/86
()ample 3-34 Miles ?un
Class Boundaries 6reuenc#
55 - ;5;5 - 5555 - $;5$;5 - $55
$55 - 3;53;5 - 355355 - ';5
$35
'3$
Below is a !reuenc# distri+ution o! milesrun per wee7 6ind the mean
Σ f @ $;
14Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
15/86
-
8/16/2019 Stats Chap03 bluman ppt
16/86
Measures o! Central "endenc#4
Median "he medianmedian is the midpoint o! the data
arra# "he s#m+ol !or the median is MD
"he median will +e one o! the data valuesi! there is an odd num+er o! values
"he median will +e the avera0e o! twodata values i! there is an even num+er o!values
16Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
17/86
Chapter 3
Data Description
.ection 3-()ample 3-'
&a0e :;
17Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
18/86
()ample 3-'4 Hotel ?ooms
"he num+er o! rooms in the seven hotels indowntown &itts+ur0h is >3, 3;;,
-
8/16/2019 Stats Chap03 bluman ppt
19/86
Chapter 3
Data Description
.ection 3-()ample 3-<
&a0e :;
19Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
20/86
()ample 3-
-
8/16/2019 Stats Chap03 bluman ppt
21/86
Measures o! Central "endenc#4
Mode "he modemode is the value that occurs most
o!ten in a data set
It is sometimes said to +e the most t#picalcase
"here ma# +e no mode, one mode8unimodal9, two modes 8+imodal9, orman# modes 8multimodal9
21Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
22/86
Chapter 3
Data Description
.ection 3-()ample 3-
&a0e :
22Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
23/86
()ample 3-4 6 .i0nin0 Bonuses
6ind the mode o! the si0nin0 +onuses o!ei0ht 6 pla#ers !or a speci!ic #ear "he+onuses in millions o! dollars are
A;, ';, 3'5, ;, 3, ;, $', ;
=ou ma# !ind it easier to sort !irst
;, ;, ;, 3, $', ';, A;, 3'5
.elect the value that occurs the most
23Bluman, Chapter 3
"he mode is ; million dollars
-
8/16/2019 Stats Chap03 bluman ppt
24/86
-
8/16/2019 Stats Chap03 bluman ppt
25/86
()ample 3-;4 Coal (mplo#ees in &*
6ind the mode !or the num+er o! coal emplo#ees percount# !or ; selected counties in southwestern&enns#lvania
;, >3, ;3, A', $;, A, >, ;3, >5$
o value occurs more than once
25Bluman, Chapter 3
"here is no mode
-
8/16/2019 Stats Chap03 bluman ppt
26/86
-
8/16/2019 Stats Chap03 bluman ppt
27/86
()ample 3-4 icensed uclear
?eactors"he data show the num+er o! licensed nuclearreactors in the 1nited .tates !or a recent 5-#earperiod 6ind the mode
;' ;' ;' ;' ;' ;> ; ; ; ;; $ ;
;' and ; +oth occur the most "he data setis said to +e +imodal
27Bluman, Chapter 3
"he modes are ;' and ;
;' ;' ;' ;' ;' ;> ; ; ; ;; $ ;
-
8/16/2019 Stats Chap03 bluman ppt
28/86
-
8/16/2019 Stats Chap03 bluman ppt
29/86
()ample 3-$4 Miles ?un per 2ee76ind the modal class !or the !reuenc# distri+utiono! miles that $; runners ran in one wee7
29Bluman, Chapter 3
"he modal class is
$;5 E $55
Class Frequency
55 E ;5
;5 E 55 $
55 E $;5 3
$;5 E $55 5
$55 E 3;5 '
3;5 E 355 3
355 E ';5 $
"he mode, the midpointo! the modal class, is
$3 miles per wee7
-
8/16/2019 Stats Chap03 bluman ppt
30/86
Measures o! Central "endenc#4
Midran0e "he midrangemidrange is the avera0e o! the
lowest and hi0hest values in a data set
2
Lowest Highest MR
+=
30Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
31/86
Chapter 3
Data Description
.ection 3-()ample 3-5
&a0e :'
31Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
32/86
()ample 3-54 2ater-ine Brea7s
In the last two winter seasons, the cit# o!Brownsville, Minnesota, reported thesenum+ers o! water-line +rea7s per month6ind the midran0e
$, 3,
-
8/16/2019 Stats Chap03 bluman ppt
33/86
Measures o! Central "endenc#4
2ei0hted Mean 6ind the weighted meanweighted mean o! a varia+le +#
multipl#in0 each value +# its
correspondin0 wei0ht and dividin0 the sumo! the products +# the sum o! the wei0hts
1 1 2 2
1 2
n n
n
wX w X w X w X X w w w w
+ + += =+ + + ∑∑
LL
33Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
34/86
Chapter 3
Data Description
.ection 3-()ample 3->
&a0e :5
34Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
35/86
()ample 3->4 Grade &oint *vera0e * student received the !ollowin0 0rades 6ind the correspondin0 G&*
35Bluman, Chapter 3
"he 0rade point avera0e is $>
wX
w X = ∑∑
Course Credits, w Grade, X
(n0lish Composition 3 * 8' points9
Introduction to &s#cholo0# 3 C 8$ points9Biolo0# ' B 83 points9
&h#sical (ducation $ D 8 point9
322.712
3 4 3 2 4 3 2 1
3 3 4 2 =× + × + × + ×
= =+ + +
-
8/16/2019 Stats Chap03 bluman ppt
36/86
&roperties o! the Mean
1ses all data values %aries less than the median or mode
1sed in computin0 other statistics, such as
the variance 1niue, usuall# not one o! the data values
Cannot +e used with open-ended classes
*!!ected +# e)tremel# hi0h or low values,called outliers
36Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
37/86
&roperties o! the Median
Gives the midpoint 1sed when it is necessar# to !ind out
whether the data values !all into the upper
hal! or lower hal! o! the distri+ution Can +e used !or an open-ended
distri+ution
*!!ected less than the mean +# e)tremel#hi0h or e)tremel# low values
37Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
38/86
&roperties o! the Mode
1sed when the most t#pical case isdesired
(asiest avera0e to compute
Can +e used with nominal data ot alwa#s uniue or ma# not e)ist
38Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
39/86
&roperties o! the Midran0e
(as# to compute Gives the midpoint
*!!ected +# e)tremel# hi0h or low values in
a data set
39Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
40/86
Distri+utions
40Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
41/86
3-$ Measures o! %ariation
How Can 2e Measure %aria+ilit#?an0e
%ariance.tandard Deviation
Coe!!icient o! %ariation
Che+#shevFs "heorem
(mpirical ?ule 8ormal9
41Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
42/86
Measures o! %ariation4 ?an0e
"he rangerange is the di!!erence +etween thehi0hest and lowest values in a data set
R Highest Lowest = −
42Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
43/86
Chapter 3
Data Description
.ection 3-$()ample 3-A
&a0e :$3$5
43Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
44/86
()ample 3-A4 Outdoor &aint
"wo e)perimental +rands o! outdoor paint aretested to see how lon0 each will last +e!ore!adin0 .i) cans o! each +rand constitute asmall population "he results 8in months9 are
shown 6ind the mean and ran0e o! each 0roup
44Bluman, Chapter 3
Brand A Brand B
; 35
-
8/16/2019 Stats Chap03 bluman ppt
45/86
()ample 3-A4 Outdoor &aint
45Bluman, Chapter 3
Brand A Brand B
; 35
-
8/16/2019 Stats Chap03 bluman ppt
46/86
Measures o! %ariation4 %ariance
.tandard Deviation "he variancevariance is the avera0e o! the
suares o! the distance each value is
!rom the mean "he standard deviationstandard deviation is the suare
root o! the variance
"he standard deviation is a measure o!how spread out #our data are
46Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
47/86
1ses o! the %ariance and .tandard
Deviation "o determine the spread o! the data
"o determine the consistenc# o! a
varia+le "o determine the num+er o! data values
that !all within a speci!ied interval in a
distri+ution 8Che+#shevFs "heorem9 1sed in in!erential statistics
47Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
48/86
Measures o! %ariation4
%ariance .tandard Deviation8&opulation "heoretical Model9 "he population variancepopulation variance is
"he population standard deviationpopulation standard deviation is
( )2
2 X N
µ σ −= ∑
( )2
X
N
µ σ
−=
∑
48Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
49/86
Chapter 3
Data Description
.ection 3-$()ample 3-$
&a0e :$5
49Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
50/86
()ample 3-$4 Outdoor &aint6ind the variance and standard deviation !or thedata set !or Brand * paint ;,
-
8/16/2019 Stats Chap03 bluman ppt
51/86
Measures o! %ariation4
%ariance .tandard Deviation8.ample "heoretical Model9 "he sample variancesample variance is
"he sample standard deviationsample standard deviation is
( )2
2
1
X X sn
−=−
∑
( ) 2
1
X X s
n
−=
−
∑
51Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
52/86
Measures o! %ariation4
%ariance .tandard Deviation8.ample Computational Model9 Is mathematicall# euivalent to the
theoretical !ormula .aves time when calculatin0 +# hand
Does not use the mean Is more accurate when the mean has
+een rounded
52Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
53/86
Measures o! %ariation4
%ariance .tandard Deviation8.ample Computational Model9 "he sample variancesample variance is
"he sample standard deviationsample standard deviation is
53Bluman, Chapter 3
( )
( )
2 2
2
1
−=
−
∑ ∑ X X n s
n n
2 s s=
-
8/16/2019 Stats Chap03 bluman ppt
54/86
Chapter 3
Data Description
.ection 3-$()ample 3-$3
&a0e :$
54Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
55/86
5A'
()ample 3-$34 (uropean *uto .ales6ind the variance and standard deviation !or the
amount o! (uropean auto sales !or a sample o! <#ears "he data are in millions o! dollars
$, , $;, $A, 3', '3
55Bluman, Chapter 3
X X2
$$;$A3''3
$5'''
-
8/16/2019 Stats Chap03 bluman ppt
56/86
Measures o! %ariation4
Coe!!icient o! %ariation"he coefficient of variationcoefficient of variation is thestandard deviation divided +# the
mean, e)pressed as a percenta0e
1se CVAR to compare standarddeviations when the units are di!!erent
100% s
CVAR X
= ×
56Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
57/86
Chapter 3
Data Description
.ection 3-$()ample 3-$5
&a0e :3$
57Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
58/86
()ample 3-$54 .ales o! *utomo+iles
"he mean o! the num+er o! sales o! cars over a3-month period is A>, and the standard deviationis 5 "he mean o! the commissions is J5$$5,and the standard deviation is J>>3 Compare
the variations o! the two
58Bluman, Chapter 3
Commissions are more varia+le than sales
5100% 5.7% ales
87CVar = × =
773100% 14.8% !"mmissi"ns5225CVar = × =
-
8/16/2019 Stats Chap03 bluman ppt
59/86
Measures o! %ariation4
?an0e ?ule o! "hum+"he Range Rule of hum!Range Rule of hum! appro)imates the standard deviation
as
when the distri+ution is unimodal andappro)imatel# s#mmetric
4
Range s ≈
59Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
60/86
Measures o! %ariation4
?an0e ?ule o! "hum+1se to appro)imate the lowestvalue and to appro)imate the
hi0hest value in a data set
60Bluman, Chapter 3
2 X s−2 X s+
#$amle: 10& 12 X Range= =
12 34
s ≈ = ( )( )
10 2 3 410 2 3 16
LOW HIGH
≈ − =≈ + =
-
8/16/2019 Stats Chap03 bluman ppt
61/86
"he proportion o! values !rom an# data set that!all within k standard deviations o! the mean will+e at least -k $, where k is a num+er 0reaterthan 8k is not necessaril# an inte0er9
Measures o! %ariation4Che+#shevFs "heorem
61Bluman, Chapter 3
: o!standard
deviations, k
Minimum &roportionwithin k standard
deviations
Minimum &ercenta0ewithin k standard
deviations
$ -'@3' >5K
3 -@A AAAK
' -5K
-
8/16/2019 Stats Chap03 bluman ppt
62/86
Measures o! %ariation4Che+#shevFs "heorem
62Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
63/86
Chapter 3
Data Description
.ection 3-$()ample 3-$>
&a0e :35
63Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
64/86
()ample 3-$>4 &rices o! Homes"he mean price o! houses in a certain
nei0h+orhood is J5;,;;;, and the standarddeviation is J;,;;; 6ind the price ran0e !orwhich at least >5K o! the houses will sell
Che+#shevFs "heorem states that at least >5K o!a data set will !all within $ standard deviations o!the mean
5;,;;; E $8;,;;;9 @ 3;,;;;
5;,;;; L $8;,;;;9 @ >;,;;;
64Bluman, Chapter 3
*t least >5K o! all homes sold in the area will have aprice ran0e !rom J3;,;;; and J>5,;;;
-
8/16/2019 Stats Chap03 bluman ppt
65/86
Chapter 3
Data Description
.ection 3-$()ample 3-$A
&a0e :35
65Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
66/86
-
8/16/2019 Stats Chap03 bluman ppt
67/86
"he percenta0e o! values !rom a data set that!all within 7 standard deviations o! the mean ina normal 8+ell-shaped9 distri+ution is listed
+elow: o! standarddeviations, 7
&roportion within 7 standarddeviations
K
Measures o! %ariation4(mpirical ?ule 8ormal9
67Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
68/86
Measures o! %ariation4(mpirical ?ule 8ormal9
68Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
69/86
3-3 Measures o! &osition
-score&ercentile
Nuartile
Outlier
69Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
70/86
Measures o! &osition4 -score
* "#score"#score or standard scorestandard score !or a valueis o+tained +# su+tractin0 the mean !romthe value and dividin0 the result +# thestandard deviation
* /-score represents the num+er o!standard deviations a value is a+ove or+elow the mean
X X s−= X
µ
σ
−=
70Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
71/86
Chapter 3
Data Description
.ection 3-3()ample 3-$
&a0e :'$
71Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
72/86
()ample 3-$4 "est .cores * student scored
-
8/16/2019 Stats Chap03 bluman ppt
73/86
Measures o! &osition4 &ercentiles
$ercentiles$ercentiles separate the data set into ;;eual 0roups
* percentile ran7 !or a datum representsthe percenta0e o! data values +elow the
datum( ), "- al(es /el" 0.5
100%*"*al , "- al(es
X !er"enti#e
+= ×
73Bluman, Chapter 3
100
n $" ×=
-
8/16/2019 Stats Chap03 bluman ppt
74/86
Measures o! &osition4 ()ample o!
a &ercentile Graph
74Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
75/86
Chapter 3
Data Description
.ection 3-3()ample 3-3$
&a0e :'>
75Bluman, Chapter 3
( l 3 3$ " t .
-
8/16/2019 Stats Chap03 bluman ppt
76/86
()ample 3-3$4 "est .cores * teacher 0ives a $;-point test to ; students
6ind the percentile ran7 o! a score o! $A, 5, $,
-
8/16/2019 Stats Chap03 bluman ppt
77/86
Chapter 3
Data Description
.ection 3-3()ample 3-3'
&a0e :'A
77Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
78/86
M ! & iti
-
8/16/2019 Stats Chap03 bluman ppt
79/86
Measures o! &osition4Nuartiles and Deciles %eciles%eciles separate the data set into ;
eual 0roups D110& D440
&uartiles&uartiles separate the data set into 'eual 0roups 125& 2MD& 375
2 median"&)i
1 median"&23 median2&)i
"he 'nterquartile Range'nterquartile Range, R 3 1
79Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
80/86
Chapter 3
Data Description
.ection 3-3()ample 3-3<
&a0e :5;
80Bluman, Chapter 3
( l 3 3< N til
-
8/16/2019 Stats Chap03 bluman ppt
81/86
()ample 3-3
-
8/16/2019 Stats Chap03 bluman ppt
82/86
Measures o! &osition4Outliers *n outlier outlier is an e)tremel# hi0h or low
data value when compared with the resto! the data values
* data value less than 1 1.5R or0reater than 3 ; 1.5R can +econsidered an outlier
82Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
83/86
3' ()plorator# Data *nal#sis
"he Five#(um!er )ummaryFive#(um!er )ummary iscomposed o! the !ollowin0 num+ers4"& 1& MD& 3& )i
"he 6ive-um+er .ummar# can +e0raphicall# represented usin0 a
Bo*plotBo*plot
83Bluman, Chapter 3
-
8/16/2019 Stats Chap03 bluman ppt
84/86
&rocedure "a+le
Constructin0 Bo)plots
6ind the !ive-num+er summar#
$ Draw a hori/ontal a)is with a scale that includes
the ma)imum and minimum data values3 Draw a +o) with vertical sides throu0h '( and
', and draw a vertical line thou0h the median
' Draw a line !rom the minimum data value to thele!t side o! the +o) and a line !rom the ma)imumdata value to the ri0ht side o! the +o)
84Bluman, Chapter $
-
8/16/2019 Stats Chap03 bluman ppt
85/86
Chapter 3
Data Description
.ection 3-'()ample 3-3A
&a0e :
-
8/16/2019 Stats Chap03 bluman ppt
86/86
()ample 3-3A4 Meteorites"he num+er o! meteorites !ound in ; 1. states
is shown Construct a +o)plot !or the data A, '>, A, A, 3A, -A35- A35