Bringing Change by Measuring Impact

Health Information Management and Applied Epidemiology for Health Care Workers in South Africa

sa-data

Module 4

Basic Numeracy in Health Care: Numeric Measures, Distributions, and Data Patterns

Learning Objectives

By the end of Module 4, you will be able to:

  • Demonstrate understanding of continuous numerical measures commonly used in health (e.g., weight, CD4 count), including selecting the appropriate units of measure, and converting common measurements among different units of measure and levels of precision (decimal places).
  • Demonstrate skills in addition, subtraction, multiplication, and division of numerical measures in health, as applied to common patient-care scenarios.
  • Define and distinguish between numerical and categorical data.
  • Using grouped data in frequency tables, explain and interpret data summaries for numerical data.
  • Explain and interpret histograms showing the distribution of numerical data.
  • List and describe the standard measures of location (mean, median, mode), spread (range), and shape (skewness). 
  • Define the concept of outliers, and explain the possible meaning(s) of outlier values.
  • Describe the concepts of normal distribution, standard deviation, and z-score; apply these concepts to measurement of child weight.
  • Using real-world numerical health data, calculate summaries of location and spread to create data summaries
  • Use tables and graphs to show distribution and central tendency of numerical data.
  • Using continuous numeric variables, describe the concept of variability of data, and its relevance in presenting summaries of health data.

This module is a bit complicated, providing you with information and activities to gain knowledge about numerical measures, mathematics, and statistics. Be sure to work slowly, at your own pace. It may help you to work through some of the activities in pairs or in groups. The information can be complex, but if you take your time to go through it, you will learn a lot!

Module 4 builds on what we learned in Module 2 about numerical measures in health care. In this module, we delve deeper, exploring how to convert and manipulate numerical measures, and how to organize and interpret data. We learn how to describe data in terms of such standard measures as mean, median, and mode, as well as range and skewness. We also learn about how to best display and interpret what the data is telling us by looking at outliers, distributions, spread, and variability.

Pre-test (10 minutes)

Part 1: Common Numerical Measures Used in Health (30 minutes)

Part 2: Quantitative & Categorical Variables (15 minutes)

Part 3: Displaying Summaries of Quantitative Measures (50 minutes)

Part 4: Measures of Central Tendency (20 minutes)

Part 5: Measures of Spread (20 minutes)

Part 6: Measures of Variability (20 minutes)

  • Summary

    In Module 4, we learnt about common quantifiable measures in a clinical setting, and built our numeracy skills even further in terms of numeric measures, distributions, and data patterns.

    In Part 1, we looked at examples from Sister Zingy’s clinic to see how she used these measures as she took care of Lebogang and other women during their ANC visits, the importance of understanding and indicating units of measure, conversion between units of measure, and using decimal points and rounding to indicate levels of precision. Basically, we learned about numerical measures commonly used in the health setting, and how to answer the question, ‘How much?’

    In Part 2, we learnt about the differences between quantitative variables (e.g., weight, haemoglobin, and blood pressure) and categorical variables (e.g., sex, marital status). 

    In Part 3, we learnt about how to display these quantitative measures, by learning how to create frequency tables, histograms, trend lines, and scatter plots. We talked about how to group variables in order to create one- and two-way frequency tables, and the difference between relative and cumulative frequency. We learned about how to display this data using histograms and bar graphs, and when to use each one to better communicate data trends. We then learned about other ways to display data through trend lines and scatter plots, and the benefits of each. 

    In Part 4, we learnt about measures of central tendency—mean, median, and mode—that can help us to better describe our data.

    In Part 5, we learnt about measures of spread—range, maximum and minimum values, percentile, and interquartile range—that can help us to better understand how values differ among patients. 

    Finally, in Part 6, we discussed measures of variability, which denote how the data is distributed around the mean or expected value. We built upon the idea of variability of measures by learning specifically how to calculate variance and standard deviation.

    In Module 5, we will talk about how we collect these data, about data systems in general, and about how to use data systems toward their intended purpose.

Post-test (10 minutes)

  • References

    Baldi, B, Moore, DS. Practice of Statistics in the Life Sciences [CD-ROM]. New York: W.H. Freeman and Company; 2009

    USAID/MEASURE Evaluation. Introduction to Basic Data Analysis and Interpretation for Health Programs Training Tool Kit. Version 1, May 29, 2011.

    Theron GB, Thompson ML: A Centile Chart for Birth Weight for an Urban Population of the Western Cape. S Afr Med J 85:1289-1292, 1995.