winslett - descriptive.ppt

advertisement
UGS 303
The
only science that
enables different experts
using the same figures to
draw different
conclusions.
What
mean?
does this really
 MEAN
 Defined
•
as the arithmetic average
Calculated by ΣX/N
 In
other words….
Add up all the raw scores in a dataset, and
then divide that number by the total number
of scores that you are given.
 MEAN
 For
example: You administered a survey to
discover people’s attitudes regarding their
college classes. You have people rate from 15 (1=extremely dislike and 5=extremely like)
how they feel about the classes they are
taking this semester.
 You
get responses of: 1,3,2,5,3,4,5,3,1,2.
 MEAN
 Add
up all your raw scores:
(1+3+2+5+3+4+5+3+1+2) = 30
 Divide this number by the total number of
scores: 30/10 = 3
 Mean = 3
 Therefore, you can conclude that most
students feel fairly indifferent about their
classes this semester.
 MEDIAN
 Defined
as the “middle score” in a set of
data
 Calculated by ordering all the raw scores in a
data set from least to greatest and taking
the middle score.
 What about when the data set presents an
even number of raw scores?? –Take the two
middle scores, add them together, and divide
by 2.
 For
the sake of time, we’re gunna use the
same data as before… 1,3,2,5,3,4,5,3,1,2.
 Order from least to greatest…
1,1,2,2,3,3,3,4,5,5.
 Now take the middle number… (since this is a
set of 10 raw scores we take the 2 middle
scores and divide by 2).
 3+3=6; 6/2= 3
 So… the median is 3 
 MODE
 Defined
as the score in a raw set of data that
occurs most frequently (there can be more
than one).
 Do
I really need to go over how to calculate
this one??
 MODE
 Using
the same scores again….
1,3,2,5,3,4,5,3,1,2 we see that the raw
score of 3 occurs most frequently.
 It
depends on what you are measuring and
what information you’re trying to convey…
 When
we hear someone talk about an
average amount, we usually think they’re
referring to the mean. But is this always
best?
 The
MEAN is largely influenced by outliers…
 For
example a data set compiled with the
following raw scores… 1,3,2,5,3,4,5,3,1,200
would have a mean of 22.7.
 Is
this a good representation of this dataset?
 Not
really…
 The
MEDIAN is not influenced by outliers…
 The
raw scores in your data set are:
1,12,28,43,56,85,110,225,576, and 984.
 Median
 Is
I
= (56+85=141… 141/2=…) 70.5
this the best representation of this data??
don’t know…
 The
MODE is not necessarily influenced by
outliers…
 2,2,4,5,6,7,8,8,9,9…
 The
you get this point.
mode for this dataset would be 2,8, & 9.
 Is
this the best representation of this
dataset?
 You
tell me…
 The
MEAN: takes every score into account
and allows for later manipulations
 The
MEDIAN: captures the center and is not
influenced by extreme scores
 The
MODE: is easy to calculate but may be
misleading
 You
can manipulate your data, and use
different measures of statistics in order to
help convey the message you want.
 This
is why 90% of statistics are a lie 
 Yes,
I am aware this is very similar to Dr.
Bias’ lecture regarding the subject, but I feel
like that’s the idea because, hell our test is
Monday!
Any Questions??
Download