graph data analysis

• Data analysis


  Turning raw data into useful information
• Purpose of data analysis

  To provide answers to questions being asked by a health program
• Even the greatest amount and best quality of data mean nothing if data are not properly analyzed - or analyzed at all
• Data analysis does not mean using a computer software package
• Data analysis


  Looking at the data in light of the questions you need to answer
• Answering program questions

  1. Compare program targets and actual program performance to learn how far you are from the targets

  2. Interpret why you have achieved or not achieved a target, and what this means for your program

  3. Answering may require more information
• Descriptive analysis

  Describes the sample/target population (demographic and clinical characteristics)<|>Does not define causality; tells you what, not why
• Basic terminology and concepts

  • Ratio
  • Proportion
  • Percentage
  • Rate
  • Mean
  • Median
  • Trend
• Central tendency

  Measures of the location of the middle or the center of a distribution of data
• Mean

  The average of your dataset<|>The value obtained by dividing the sum of a set of quantities by the number of quantities in the set
• Calculating the mean

  Sum of all values / Number of values
• Median

  The middle of a distribution (when numbers are in order: that is, half of the numbers are above the median and half are below the median)<|>The median is not as sensitive to extreme values as the mean
• Calculating the median

  1. For odd number of values, median = the middle number

  2. For even number of values, median = mean of the two middle numbers
• Mean vs Median


  Use mean when data is normally distributed<|>Use median when data has outliers or is skewed
• Trend

  A pattern of gradual change in a condition, output, or process, or an average or general tendency of a series of data points to move in a certain direction over time, represented by a line or curve on a graph
• Calculating trends

  Observe patterns of change over time
• Key messages

  Purpose of analysis: Provide answers to programmatic questions<|>Descriptive analyses describe the sample or target population<|>Descriptive analyses do not define causality. That is, they tell you what, not why
• Types of charts

  • Column/bar
  • Circular area
  • Line
  • Scatter plot
  • Bullet
  • Pie
  • Stacked bar
  • Stacked column
  • Area
  • Dual-axis line
  • Bubble
• 5 questions to ask when choosing a chart

  1. Want to compare values?

  2. Want to show the composition of something?

  3. Want to understand the distribution of your data?

  4. Interested in analyzing trends in your data set?

  5. Want to better understand the relationships among value sets?
• Column


  • To show a comparison among different items
  • To show a comparison of items over time
• Bar


  • Should be used to avoid clutter when one data label is long or if you have more than 10 items to compare
  • Can also be used to display negative numbers
• Line


  • A line chart reveals trends or progress over time
  • Can be used to show many different categories of data
  • Use a line chart to show a continuous data set
• Dual axis

  • Used with 2-3 data sets, at least one of which is based on a continuous set of data, and another of which is better suited to being grouped by category
  • Should be used to visualize a correlation, or the lack thereof, between these three data sets
• Area

  • Useful for showing part-to-whole relationships, such as individual data's contribution to the total for a given period
  • Helps you analyze both overall and individual trend information
• Stacked bar


  • Should be used to compare many items and show the composition of each one
  • Represents components of a whole and compares wholes
• Pie

  • Represents percentages, with the segments totaling 100
• Scatter plot

  • Can show relationship between two variables, or reveal the distribution trends
  • Should be used when there are many data points, and you want to highlight similarities in the data set
  • Useful when you are looking for outliers or want to understand the distribution of your data