NAME:   Readability of Educational Materials for Patients with Cancer
TYPE:   Two independent samples
SIZE:   Two variables, one with 30 observations and the other with 63.
	The data file contains frequency distributions for the two
	variables over a common range of grade levels.

DESCRIPTIVE ABSTRACT:
The readabilities of 30 pamphlets about cancer are compared to the
reading comprehension levels of 63 patients with cancer.  Both
variables are measured in grade levels.  The data are presented as
frequencies of occurrence over grade levels for both the pamphlet
readabilities and the reading levels of the patients.

SOURCE:
Cooley, M. E., Moriarty, H., Berger, M. S., Selm-Orr, D., Coyle, B., and
Short, T. (1995), "Patient Literacy and the Readability of Written
Educational Materials for Patients with Cancer," to appear in _Oncology
Nursing Forum_, October 1995.

VARIABLE DESCRIPTIONS:
Columns
1 - 2   Grade level 
4 - 5   Frequency of occurrence for brochure readabilities
7 - 8   Frequency of occurrence for patient reading levels

Values are aligned and delimited with blanks.

SPECIAL NOTES:
One key to the analysis of these data is the ordinal nature of the
grade level data for patient reading levels.  In particular, the
extreme values of the WRAT R2 scale are "2", which really means "less
than third grade," and "13", which really means "College and above."

STORY BEHIND THE DATA:
A study conducted on patients at the Philadelphia Veterans Affairs
Medical Center compared the readabilities of brochures informing
patients about cancer and its treatment to the reading levels of
patients with cancer.  The purpose of the data collection and analysis
was to determine if existing written materials about cancer are aimed
at a reading level too far above the comprehension levels of the
patients.

Although analysts are tempted to consider means and medians to make
comparisons, the clearest conclusions are reached by the interpretation
of simple graphical displays such as histograms.

PEDAGOGICAL NOTES:
The dataset in this example is small, but it can be used to initiate
discussion of many statistical issues.  The sample of patients was
self-selected, which may lead to data collection issues.  The methods
for assigning grade levels to both readability and patient literacy
could be discussed.

Perhaps more importantly, the scale of measurement for the grade levels
affects choice of measures of center, and the distribution of the data
values calls into question whether measuring center even makes sense in
this case!

Finally, the clarity of the histograms and the resulting
interpretations should be used to highlight the value of simple
graphical displays.

Additional information on the pedagogical uses of these data can be
found in the "Datasets and Stories" article "Readability of Educational
Materials for Patients with Cancer" in the _Journal of Statistics
Education_ (Short, Moriarty, and Cooley 1995).  

SUBMITTED BY:
Thomas H. Short                                 
Department of Mathematical Sciences             
Villanova University                            
Villanova, PA 19085                             
short@monet.vill.edu

Helene Moriarty 
Philadelphia Veterans Affairs Medical Center
University and Woodland Avenue
Philadelphia, PA 19104                         
and 
College of Nursing
Villanova University
Villanova, PA 19085

Mary E. Cooley
Philadelphia Veterans Affairs Medical Center
University and Woodland Avenue
Philadelphia, PA 19104