Why this is useful Failure as a statistician/ analyst often is failure to clearly communicate Need...

29
Why this is useful Failure as a statistician/ analyst often is failure to clearly communicate Need to communicate results to non- technical decision-makers – politicians, judges Can provide insight into data for both internal (YOU) and external uses

Transcript of Why this is useful Failure as a statistician/ analyst often is failure to clearly communicate Need...

Page 1: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Why this is useful

Failure as a statistician/ analyst often is failure to clearly communicate

Need to communicate results to non-technical decision-makers – politicians, judges

Can provide insight into data for both internal (YOU) and external uses

Page 2: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Descriptive uses of categorical data

Describe sample Check Data Quality Answer Descriptive Questions

Page 3: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Please pay attention ….Categorical data can be either nominal or ordinal. It is perfectly reasonable to discuss whether ordinal data are skewed (and often, one of the most interesting findings is that it is)With categorical data, it is also useful to look at distributions

Page 4: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Questions related to distributionsWhat is the distribution of students’ expectations about their own likely educational attainment?

Is the distribution of students’ expectations about their own likely educational attainment skewed?

Page 5: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Easy creation of charts & tables with SAS Enterprise Guide Just pointing and clicking Also available free to university researchers

and students via SAS on-demand Characterize data tasks gives frequency

distributions for all categorical variables and charts for ALL variables

Page 6: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Answer the questions

The distribution of students’ expectations for educational attainment are shown above. The median expectation was to finish a Bachelor’s degree.

Only 17.1% of students expected to complete less than a four-year degree.

The distribution of educational expectations is very skewed.

Page 7: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Categorical data that is in order

The distribution of homework hours is somewhat positively skewed

Mean = 2.68Median = 1-2 hours ( Category 3)Mode = 2

Page 8: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Getting the data

Figure 1.1

FILE> OPEN> DATA

Page 9: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Tasks> describe> characterize data

Figure 1.5

ALWAYS DO THIS !!

Page 10: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Just click through the windows and accept all of the defaults.

Page 11: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-
Page 12: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Some Coding

ODS GRAPHICS ON ; *PROC FREQ DATA = dsname ;

TABLES varname ; Will produce histograms and one-way tables

* Not needed in SAS 9.3

Page 13: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

(Sounds more impressive than it is)

Bi-variate categorical data analysis

Page 14: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Homes without computers have fewer books

Page 15: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Children of mother’s with more education are less likely to fail

Notice how the furtherdown the column yougo, the smaller thecolumn percentages

Page 16: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Bi-variate distributions Is there a relationship between school failure

and mother’s education?

Is there a relationship between the number of books (this was a category) in the home and whether a family has a computer

Is there a relationship between mother’s education and father’s education

Page 17: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Answer by trend, proportion, odds The trend of the data showed a lower

likelihood of a student failing a grade the higher the educational level of the moms surveyed.

At the 0-11 yrs of education level 73% of the students never failed a grade while at 16 yrs or more it was 88% never having failed.

Students whose mothers had not finished high school were more than twice as likely to fail a grade as children of college graduate mothers

Page 18: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Mothers tend to be married to fathers with similar education

Note that the highestrow percentages tendto be in the diagonalswhere the parents’ education is the same

Page 19: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Some More CodingPROC FREQ DATA = dsname ;

TABLES varname1 * varname2 / AGREE ;

FOR CORRELATED DATA

Page 20: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Correlated Data

Page 21: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

McNemar’s Test

Page 22: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Correlated Data

Page 23: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Cohen’s Kappa

1.0 = perfect agreementNegative Kappa is not an error, it means the two agree less than chance

= Probability observed – Probability expected 1 – Probability expected

Page 24: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Chi-square (wrong)

Notice you do NOT get identical p-value

Page 25: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Fisher (wrong)

Notice you do NOT get identical p-value

Page 26: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Some More CodingPROC FREQ DATA = dsname ; TABLES varname1 * varname2 / chisq ;

Page 27: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Chi-square (right)

Page 28: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Right

Page 29: Why this is useful  Failure as a statistician/ analyst often is failure to clearly communicate  Need to communicate results to non-technical decision-

Right