You are on page 1of 2

Fab Five Data Analysis In-Class Activity Week 9

A researcher is measuring the systolic blood pressure of a sample of


African-American males in their forties. She is using a pretest-posttest
design to test the effectiveness of an herbal supplement that is
believed to be a mild antihypertensive at lowering the blood pressure
of middle-aged African-American men with mild hypertension over a
six-week period. The dataset below represents the pretest scores:

1. Is this data nominal, ordinal, interval, or ratio? Explain why.


This data is ordinal because it can be placed order but not ratio or
interval because there are not equal intervals between the values.
2. Calculate the mean
1,570/10=157
3. Identify the median
130+130=260/2=130
4. Identify the mode
130
I5. dentify which group has the higher standard deviation: Patients 1-5
or Patients 6-10. Explain why you are able to answer this question
without doing any calculations.
Patients 6-10 would have the higher standard deviation due to the
range being 120-300 and for patients 1-5 only having a range of 120175.
6. Identify the outlier in this dataset.
300
7. Now, remove the outlier from this dataset and recalculate the mean,
median, and mode. How have these values changed? Based on how
these values have changed, explain how outliers can affect your data.
When we remove outliers we are changing the data, it is no longer
"pure", so we shouldn't just get rid of the outliers without a good
reason. In regards to this data it seems that outliers have the biggest
effect on the mean, and not on the median or mode. The mean went
down slightly, but the median and mode stayed the same.
Mean=141.1
Median=130

Mode=130
8. Do you believe it would be justified for this researcher to remove the
outlier from the analysis?
An outlier may be due to variability in the measurement or it may
indicate experimental errors and may sometimes be excluded from the
data set. Outlier data are not always 'errors' (e.g. resulting from
experimental artifacts or typing errors in data files), but just the result
of an unusual event or factor that may have been missed during the
study. A normal systolic range is less than 120, pre hypertension 120139, stage 1 hypertension is 140-159, stage 2 hypertension 160 or
higher, and hypertensive crisis 180 or higher. In this study with or
without the outlier the mean will still fall in a hypertension stage and
would make it not justifiable to remove the outlier.
9. Discuss with your group members the arguments for and against
removing outliers from your data.
Removing the outliers from our data would not show a full picture of
the data we collected. The argument for removing the outliers could
include that it doesnt change the overall trend in the data.
10. Using Excel or other graphics software, convert these data into a
histogram. Do you believe that the histogram provides more clarity
than the chart shown above? Why or why not?
The histogram does show more clarity than the above chart. It provides
a better visual of the frequency of each blood pressure rather than just
listing it in a table.

Histogram
Frequency

You might also like