To Activities

IT 121 -- Activity A6

Measurement Error

What to submit (on COL): Your Excel file containing the data analysis for this activity. Whenever appropriate, write explanations right into your Excel file (I want explanations). When you give explanations, write them in text boxes or callouts. Make everything look nice. Name your Excel file group1.xls or group2.xls, or ..., depending on the number of your group. In your Excel file be sure to include the names of all group members (at the top of the file).

1. For this activity, work in groups of four (timer, recorder, counter, reader).

2. For this activity, one observation consists of timing a subject count silently to 30 seconds with his or her eyes closed.

   Use this online stopwatch for timing.

   Counting to 30 seconds is a form of measurement; you are investigating the measurement error.

3. Each subject should repeat the timing 10 times. This will give a total of 40 timings for each group.

4. Enter your data into an Excel spreadsheet. Everything (raw data, analysis, and explanations) should be in just one sheet - call the sheet Time Measures.

5. Compute these descriptive statistics: mean, standard deviation, Q0, Q1, Q2, Q3, Q4.

6. Create the histogram of the timing data. Make sure you label the x and y axis and the overall histogram (what's it a histogram of?). Interpret your histogram.

7. See the Excel file NBS-10.xls for examples of computing descriptive statistics and creating histograms.

8. Answer these questions:
   1. Estimate the standard deviation from the minimum and maximum by the rule of thumb:
      Estimated SD = (max - min) / 4.

      Compare this estimate with the actual SD.

   2. Are there any outliers that appear to throw off your estimate in part a? Explain. Use the definition of outliers using IQR, Q1 and Q3 in the Outliers section of the 4/21 lecture notes.

   3. Does there appear to be any systematic bias in your data? Explain.