Module 4 - Carolina Population Center

advertisement
DATA ANALYSIS
Module 4
Part 1 – Key Concepts
Learning Objectives
 Understand the definition and purpose of data
analysis
 Define statistical and M&E key concepts in data
analysis
Data Analysis
 Turning raw data into useful information
 Purpose is to provide answers to questions being
asked at a program site or research questions
 Even the greatest amount and best quality data
mean nothing if not properly analyzed—or if not
analyzed at all
Data Analysis
 Analysis does not mean using computer software
package
 Analysis is looking at the data in light of the
questions you need to answer:
 How would you analyze data to determine: “Is
my program meeting its objectives?”
Answering programmatic questions
 Question: Is my program meeting its objectives?
 Analysis: Compare program targets and actual
program performance to learn how far you are from
target.
 Interpretation: Why you have or have not achieved
the target and what this means for your program.
 May require more information.
Descriptive analysis
 Describes the sample/target population
(demographic & clinic characteristics)
 Does not define causality – tells you what, not
why
 Example – average number of clients seen
per month
Basic terminology and concepts
 Statistical terms
 Ratio
 Proportion
 Percentage
 Rate
 Mean
 Median
Ratio
 Comparison of two numbers expressed as:
 a to b, a per b, a:b
 Used to express such comparisons as clinicians
to patients or beds to clients
 Calculation a/b
 Example – In district X, there are 600 nurses and
200 clinics. What is the ratio of nurses to clinics?
600
= 3 nurses per clinic, a ratio of 3:1
200
Calculating ratios
 In Kwakaba district, there are 160 nurses and 40
clinics
 What is the nurse-to-clinic ratio?
160
40
=4
4:1 or 4 nurses to 1 clinic
Proportion
 A ratio in which all individuals in the numerator are
also in the denominator.
 Used to compare part of the whole, such as proportion
of all clients who are less than 15 years old
 Example: If 20 of 100 clients on treatment are less
than 15 years of age, what is the proportion of young
clients in the clinic?
 20/100 = 1/5
Calculating proportions
 Example: If a clinic has 12 female clients and 8
male clients, then the proportion of male clients
is 8/20, or 2/5
 12+8 = 20
 8/20
 Reduce this, multiple of 4 = 2/5 of clients = male
Percentage
 A way to express a proportion (proportion
multiplied by 100)
 Expresses a number in relation to the whole
 Example: Males comprise 2/5 of the clients, or
40% of the clients are male (0.40 x 100)
 Allows us to express a quantity relative to
another quantity. Can compare different groups,
facilities, countries that may have different
denominators
Rate
 Measured with respect to another measured
quantity during the same time period
 Used to express the frequency of specific events
in a certain time period (fertility rate, mortality
rate)
 Numerator and denominator must be from same
time period
 Often expressed as a ratio (per 1,000)
Source: U.S. Census Bureau, International Database.
Infant Mortality Rate
• Calculation
• # of deaths ÷ population at risk in same time
period x 1,000
• Example – 75 infants (less than one year) died
out of 4,000 infants born that year
• 75/4,000 = .0187 x 1,000 = 18.7
19 infants died per 1,000 live births
Calculating mortality rate
In 2009, Mondello clinic had 31,155 patients on
ART. During that same time period, 1,536 ART
clients died.
1,536
= .049 x 1,000 = 49
31,155
49 clients died
(mortality rate) per
1,000 clients on
ART
Rate of increase
 Calculation
 Total number of increase ÷ time of increase
 Used to calculate monthly, quarterly, yearly
increases in health service delivery. Example:
increase in # of new clients, commodities
distributed
 Example: Condom distribution in Jan. = 200; as of
June = 1,100. What is the rate of increase?
 1,100 - 200 = 900/6 = 150 (150 condoms per mo)
Calculating rate of increase
In Q1, there were 50 new FP users, and in Q2
there were 75. What was the rate of increase from
Q1 to Q2?
Example: 75 - 50 = 25 /3 = 8.33 new clients/mo
Central tendency
Measures of the location of the middle or the center
of a distribution of data
 Mean
 Median
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
 Example: (22+18+30+19+37+33) = 159 ÷ 6 =
26.5
 The mean is sensitive to extreme values
Calculating the mean
 Average number of clients counseled per month
–
–
–
–
–
–
January: 30
February: 45
March: 38
April: 41
May: 37
June: 40
(30+45+38+41+37+40) = 231÷
6 = 38.5
Mean or average = 38.5
Median
 The middle of a distribution (when numbers are in
order: 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
 Odd number of numbers, median = the middle number
 Median of 2, 4, 7 = 4
 Even number of numbers, median = mean of the two
middle numbers
 Median of 2, 4, 7, 12 = (4+7) /2 = 5.5
Calculating the median
 Client 1 – 2
 Client 2 – 134
 Client 3 – 67
 Client 4 – 10
 Client 5 – 221
 = 67
 = 67+134 = 201/2 = 100.5
Use the mean or median?
CD4 count
Client 1
9
Client 2
11
Client 3
100
Client 4
95
Client 5
92
Client 6
206
Client 7
104
Client 8
100
Client 9
101
Client 10
92
Key messages
 Purpose of analysis is to provide answers to
programmatic questions
 Descriptive analyses describe the sample/target
population
 Descriptive analyses do not define causality –
that is, they tell you what, not why
Part 2: Basic analyses
Part 2: Learning Objectives
 Identify approaches for setting targets
 Understand common analyses that calculate
program coverage and retention
 Calculate program coverage and retention
Terminology
 Indicator
 Target
 Program coverage
 Service availability
 Service utilization
 Program retention
Indicator
 Program element that needs tracking
 Measures an aspect of a program’s performance
 Measures changes over a period of time
• # of new family planning users
• # of clients currently on ART
 Expressed as a number or percentage
Target Definition
 A specified level of performance for a
measure (indicator), at a predetermined
point in time (i.e., achieve ‘x’ by ‘y’ date)
 Overall target
 Annual targets
Why Set Targets?
 Targets help program staff with:
 Planning
– Staffing and service delivery
– Commodities
 Monitoring progress
– Break long-term goals into manageable pieces
– Check progress on indicators
Setting Reasonable Targets
 The range of values for a given indicator
can be from 0% to 100%.
• Example: The theoretical range for the Polio
indicator is between 0% of children
immunized (bad) and 100% immunized
(ideal)
• Is it appropriate to set the Polio indicator
target at 100% for a given program?
Why/why not?
Setting Reasonable Targets
 Example: In Somalia, the national CPR from
2007 to 2009 was15%. The following year, a
national target was set for 70%.
 Is it appropriate to set the CPR target for
Somalia at 70%? Why/why not?
Overall Target Setting Approaches
 There are three approaches to set a target :
 Established long-term goals by contacting that
national program
 Past performance (of your program, increasing by
no more than 10%)
 Local high performer (a stellar program nearby)
 Consider the number of clients your program can
realistically expect to serve during a given period
of time
Annual Target Setting
 Determine the increase your program needs
to gain to reach your overall target
 Divide that number by the number of years in
which you would like to achieve the target
 Add the number to your baseline indicator for
each year
Considerations for Target Setting
 Ensure you have an agreed-upon and realistic
definition of target population
 Set a realistic target to achieve in the long term
and short term
Importance of Defining the Target
Population: Case Example
 Target was 372 children to be immunized
 Actual was 488 children immunized
 To calculate the % target achieved, use
(Actual/Target) * 100
 488/372 = 1.31*100 = 131%
 How could the clinic have surpassed its
target by so much?
Implications of Incorrect Target
Setting: Case Example
 You don’t really know to what extent you’re fully
immunizing the children in your setting
 If your program purchases commodities (e.g.,
vaccines) based on the target set, supply could
run out
 If you set your target too low, you may not have
enough vaccines, leading to disease outbreaks
Common Analyses
 Program Coverage
 Extent to which a program reaches its intended
target population, institution, or geographic area
 Compare current performance to prior year/quarter
 Compare performance between sites
 Program Retention
 Extent to which the range of services is being
delivered as initially intended so that client dropouts are minimal
Why do we need to measure
coverage?
 To understand program progress
 To determine if the target is reached
 Clients, commodities, adherence…
 To determine if one target is reached more
effectively than another
• Are there underserved area/regions, subpopulations?
39
Program coverage
 Extent to which a program reaches its intended target
population, institution, or geographic area
 Utilization:
 Is the target population utilizing services, accessing
commodities, being reached with services?
 Availability:
 Are the services available where there is a need?
Utilization calculation
Percentage of the target population utilizing
services
# of individuals in target population
using a service
------------------------------------------# of individuals in target population
X 100
Utilization calculation: Example
 No. of persons educated as of 6/12/09 = 300
 Goal for 12/31/09 = 900
300
900
= 0.33 x 100 =
33%
 You have reached 33% of your target group with
education messages
Comparison of time periods
 Compare percentage achieved toward target for
different time periods, different sites, etc.
 Rate of increase
 As of January, 70 people educated; by June, 300
people
 300 – 70 = 230 increase in people educated
 230/6 = 38.3 new people educated per month
over the 6 months
Utilization of PMTCT Programs
All pregnant women
(2,000)
PMTCT
Target
(1,000)
Target population
Sought prenatal care
(600)
Utilization =
Utilization =
Service users
600/1,000 = 0.6
Counseled &
Tested for HIV (500)
0.6 x 100 = 60%
Program coverage
 Extent to which a program reaches its intended target
population, institution, or geographic area
 Utilization:
 Is the target population utilizing services, accessing
commodities, being reached with services?
 Availability:
 Are the services available where there is a need?
Availability calculation
 Number of service outlets available per target
population
 # of clinics with PMTCT per # of pregnant women
 Expressed as a ratio
PMTCT clinic availability
 There are 8 clinics offering PMTCT & 100,000
pregnant women in region X.
 Ratio of clinics to pregnant women 8:100,000
 Reduce to (1:12,500) pregnant women
 The standard recommendation is 1 clinic with
PMTCT services per 10,000 pregnant women
 Clinic availability is not reaching the target
Availability + Utilization = Coverage
 Service availability is 1:12,500
 Service availability target is 1:10,000
 PMTCT service utilization is 25% off the target
 What can we conclude?
 Service availability and utilization are too low; the
program is not meeting the needs of pregnant
women.
Program retention
 Measures if the range of services are being delivered
as initially intended
 Determines program retention, i.e., is the project
keeping clients through entire package of services?
• Important in clinical programs where drug adherence is
an issue (TB, HIV/AIDS, immunization) and there are
multiple steps (PMTCT)
49
Retention example: Immunization
Utilization
Enter service
Polio dose 1
Completion
Polio dose 2
Polio dose 3
All pregnant women
(2,000 women)
PMTCT
Target
(1,000)
PMTCT Program Retention
Sought prenatal care (600)
350 received HIVresult or no result
Tested for HIV (500)
40 received
prophylaxis
100 received HIV+
result
All pregnant women
(2,000 women)
PMTCT Program Retention
1,000
Sought prenatal care
500
Tested for HIV
40 received
prophylaxis
350 received HIVresult
100 received HIV+
result
All pregnant women
(2,000 women)
PMTCT
Target
(1,000)
PMTCT Program Retention
Sought prenatal care (600)
350 received HIVresult
Tested for HIV (500)
40 received
prophylaxis
100 received HIV+
result
All pregnant women
(2,000 women)
PMTCT
Target
(1,000)
PMTCT Program Retention
Sought prenatal care (600)
350 received HIVresult or no result
Tested for HIV (500)
40 received
prophylaxis
100 received HIV+
result
All pregnant women
(2,000 women)
PMTCT
Target
(1,000)
PMTCT Program Retention
Sought prenatal care (600)
350 received HIVresult or no result
Tested for HIV (500)
40 received
prophylaxis
100 received HIV+
result
Key messages
 Target Setting – A specified level of performance
for a measure (indicator) at a predetermined point
in time. Both overall and annual targets are set
 Coverage – extent to which a program reaches its
intended target population, institution, or
geographic area
 Retention – the extent to which the range of
services are being delivered as initially intended,
with clients retained throughout the full package of
services
Part 3: Data Presentation and
Interpretation
Part 3: Learning Objectives
 Understand different ways to best summarize
data
 Choose the right table/graph for the right data
 Interpret data to consider the programmatic
relevance
Summarizing data
 Tables
 Simplest way to summarize data
 Data are presented as absolute numbers or
percentages
 Charts and graphs
 Visual representation of data
 Data are presented as absolute numbers or
percentages
Basic guidance when
summarizing data
 Ensure graphic has a title
 Label the components of your graphic
 Indicate source of data with date
 Provide number of observations (n=xx) as a
reference point
 Add footnote if more information is needed
Tables: Frequency distribution
Set of categories with numerical counts
Year
Number of births
1900
61
1901
58
1902
75
Tables: Relative frequency
number of values within an interval
total number of values in the table
Year
x 100
# births (n)
Relative frequency (%)
1900–1909
35
27
1910–1919
46
34
1920–1929
51
39
Total
132
100.0
Tables
Percentage of births by decade between 1900 and 1929
Year
Number of births
(n)
Relative frequency
(%)
1900–1909
35
27
1910–1919
46
34
1920–1929
51
39
Total
132
100.0
Source: U.S. Census data; 1900–1929.
Charts and graphs
 Charts and graphs are used to portray:
 Trends, relationships, and comparisons
 The most informative are simple and selfexplanatory
Use the right type of graphic
 Charts and graphs
 Bar chart: comparisons, categories of data
 Line graph: display trends over time
 Pie chart: show percentages or proportional
share
Bar chart
Comparing categories
6
5
4
Site 1
3
Site 2
Site 3
2
1
0
Quarter 1
Quarter 2
Quarter 3
Quarter 4
% o f new enrollees tested for
HIV
Percentage of new enrollees tested for HIV at each
site, by quarter
6
5
4
3
2
Site 1
Site 2
Site 3
1
0
Quarter 1
Q1 Jan–Mar
Quarter 2
Quarter 3
Q2 Apr–June
Q3 July–Sept
Months
Quarter 4
Q4 Oct–Dec
Data Source: Program records, AIDS Relief, January 2009 – December 2009.rce:
Quarterly Country Summary: Nigeria, 2008
Has the program met its goal?
% of new enrollees tested
for HIV
Percentage of new enrollees tested for HIV at each site, by
quarter
60%
50%
40%
30%
Site 1
Site 2
Site 3
20%
10%
Target
0%
Quarter 1
Quarter 2
Quarter 3
Quarter 4
Data Source: Program records, AIDS Relief, January 2009 – December 2009.. uarterly
Country Summary: Nigeria, 2008
Stacked bar chart
Represent components of whole & compare wholes
Number of Months Female and Male Patients Have Been
Enrolled in HIV Care, by Age Group
Females
4
10
0-14 years
15+ years
Males
3
6
0
5
10
15
Number of months patients have been enrolled in HIV care
Data source: AIDSRelief program records January 2009 - 20011
Line graph
Displays trends over time
Number of Clinicians Working in Each Clinic During Years 1–4*
Number of clinicians
6
5
4
Clinic 1
3
Clinic 2
2
Clinic 3
1
0
Year 1
*Includes doctors and nurses
Year 2
Year 3
Year 4
Line graph
Number of Clinicians Working in Each Clinic During Years 1-4*
6
Number of clinicians
5
4
Clinic 1
3
Clinic 2
Clinic 3
2
1
0
Y1 1995
1
Year
Y2Year
19962
Y3Year
19973
Zambia Service Provision Assessment, 2007.
*Includes doctors and nurses
Y4
1998
4
Year
Pie chart
Contribution to the total = 100%
Percentage of All Patients Enrolled by Quarter
8%
10%
1st Qtr
2nd Qtr
23%
N=150
3rd Qtr
59%
4th Qtr
Interpreting data
Interpreting data
 Adding meaning to information by making
connections and comparisons and exploring
causes and consequences
Relevance
of finding
Reasons
for finding
Consider
other data
Conduct
further
research
Interpretation – relevance of finding
 Adding meaning to information by making
connections and comparisons and exploring
causes and consequences
Relevance
of finding
Reasons
for finding
Consider
other data
Conduct
further
research
Interpretation – relevance of finding
 Does the indicator meet the target?
 How far from the target is it?
 How does it compare (to other time periods,
other facilities)?
 Are there any extreme highs and lows in the
data?
Interpretation – Possible causes?
• Supplement with expert opinion
• Others with knowledge of the program or target
population
Relevance
of finding
Reasons
for finding
Consider
other data
Conduct
further
research
Interpretation – Consider other data
Use routine service data to clarify questions
- Calculate nurse-to-client ratio, review
commodities data against client load, etc.
Use other data sources
Relevance
of finding
Reasons
for finding
Consider
other data
Conduct
further
research
Interpretation – Other data
sources
 Situation analyses
 Demographic and health surveys
 Performance improvement data
Relevance
of finding
Reasons
for finding
Consider
other data
Conduct
further
research
Interpretation – conduct further
research
 Data gap
conduct further research
 Methodology depends on questions being asked
and resources available
Relevance
of finding
Reasons
for finding
Consider
other data
Conduct
further
research
Key messages
 Use the right graph for the right data
 Tables – can display a large amount of data
 Graphs/charts – visual, easier to detect patterns
 Label the components of your graphic
 Interpreting data adds meaning by making
connections and comparisons to program
 Service data are good at tracking progress &
identifying concerns – do not show causality
Activity 5: Calculating
coverage and retention
Learning Objectives
 Use basic statistics to measure coverage and
retention
 Develop graphs that display performance
measures (utilization, trends)
 Interpret performance measures for
programmatic decision making
Small group activity
 Form groups of 4–6
 Each group reviews 2 worksheets from Excel file
and answers the questions (1 hr 45 min)
 Each group presents 2 findings from each
worksheet, focusing on the programmatic
relevance of the findings (10 min per group)
 Audience provides feedback on analysis and
interpretation (notes errors, additional
interpretation) (10 min per group)
THANK YOU!
MEASURE Evaluation is a MEASURE project funded by the U.S.
Agency for International Development and implemented by the
Carolina Population Center at the University of North Carolina at
Chapel Hill in partnership with Futures Group International, ICF Macro,
John Snow, Inc., Management Sciences for Health, and Tulane
University. Views expressed in this presentation do not necessarily
reflect the views of USAID or the U.S. Government. MEASURE
Evaluation is the USAID Global Health Bureau's primary vehicle for
supporting improvements in monitoring and evaluation in population,
health and nutrition worldwide.
Visit us online at http://www.cpc.unc.edu/measure.
Download