RESEARCH DISCIPLINE
Public Health Project Data Analysis
Health Indicators From Public Datasets
Public health surveillance data analysis turns surveillance reports, vital statistics and national surveys into indicators that are clearly defined, comparable across sources and years, and shown with the intervals that a program reviewer needs to read them.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
What Goes Into Building a Public Health Indicator?
Public health project data analysis supports projects built on indicators from public sources such as surveillance reports, vital statistics and national surveys. The analysis starts with the indicator itself: its numerator, denominator, population and years, because two datasets can give the same name to different quantities. Places, groups or periods are compared only after each source has been read for how its figures were produced. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides this support: an in-house PhD-level statistician builds the indicators, carries out the comparisons and explains the limits of each source. Recommendations for policy are drawn by the researcher, not by the analysis.
WHEN TO ENGAGE
Public Data Projects That Need an Analysis Plan
Public data look ready for analysis, yet comparing two sources or two years raises questions that the documentation alone does not settle.
Indicators From Several Agencies Indicators are taken from different agencies, and their definitions, age groups and reference years have to be reconciled before they are placed side by side.
Trend Across Changed Definitions A series spans a revision in coding or survey method, and the project must decide whether the years before and after can be compared at all.
Linking Files at Place Level Case counts from one source are divided by population from another, and the geographic units and years of the two must match before a rate is computed.
Individual Records in a Public File A public-use microdata file supports a model of an individual outcome, and the project must choose the model, the covariates and the way estimates are reported.
WHAT THE ANALYSIS COVERS
How a Public Indicator Is Built and Compared
The analysis is organized around the indicator, beginning with what the source measured and ending with how the comparison is shown.
Source Review
The documentation of each dataset is read for coverage, definitions, suppression rules and known breaks, and a summary of what was found is kept with the analysis.
Indicator Construction
Numerators, denominators and age groups are defined and applied consistently, and prevalence or incidence is estimated following Incidence & Prevalence Analysis.
Place and Group Comparisons
Places or groups are compared by differences and ratios of the indicator, each with an interval that reflects whether the figure is a full count or a survey estimate.
Comparability Over Time
Years are compared only where definitions and methods allow it, as when case definitions change in a notifiable disease trend analysis, and a break in the series is marked on the figure and not smoothed over.
Individual-Level Models
For binary outcomes in microdata, such as current smoking, Logistic Regression estimates adjusted associations, with the survey design taken into account where the file supplies it.
Reporting the Indicators
Findings are presented as Public Health reporting expects, with the sources, years and suppressions listed, within the broader field of Medical, Health & Clinical Research.
COMMON INDICATOR ISSUES
Comparison Errors Built Into Public Figures
Each fault arises from treating a published number as a fixed quantity, when the number depends on how its source produced it.
Same Name, Different Quantity Two sources report a rate that differs in its denominator or its age range, and the figures are plotted together as though they measured the same thing.
Mismatched Years in Numerator and Denominator Case counts for one year are divided by a population estimate for another, or by a population estimate from a different release, which shifts the rate for reasons unrelated to health.
Crude Rates Across Age Structures Places with older populations show higher crude rates for many outcomes, so comparing them without standardization mistakes age structure for risk.
Survey Estimates Set Beside Counts A prevalence from a national survey is placed beside a registered count as if both were exact, though the survey figure carries sampling error and the count does not.
Outcome Defined by Available Variable The closest variable in a public file is taken as the outcome, although its question wording measures something narrower than the concept it names.
Latest Year Taken as Final The most recent release of an indicator is used as if settled, though reporting delay in surveillance data leaves recent counts incomplete and agencies often revise recent estimates, so a finding can shift after publication.
DELIVERABLES
Source-to-Table Files for Each Indicator
The records trace each figure from its public source to its place in a table, so the project can be audited and updated when a source is revised.
Source Register Each dataset, its release or version, the date accessed and the definitions that matter, in a table that supports the methods section.
Indicator Definitions The numerator, denominator, population and years of every indicator, with the standard population used for any standardized rate.
Indicator Estimates by Group Indicators for each place, group or period with 95% confidence intervals, the counts behind them and flags for suppressed values.
Series Figures Trend figures that mark breaks in definition and show intervals, in the layout your program or journal requires.
Microdata Model Results Adjusted estimates with intervals for any individual-level model, with a note on how the survey design was handled.
Update Commands The commands that rebuild every table from the downloaded files, so the analysis can be refreshed when a source is revised.
ILLUSTRATIVE EXAMPLE
Immunization Coverage Across a Revised Denominator
A break in a public data series, found and marked before any county was compared.
ILLUSTRATIVE EXAMPLE · CHILDHOOD IMMUNIZATION COVERAGE RESEARCH · UNITED STATES
A Master's researcher in public health compared childhood immunization coverage across counties over a decade, using coverage figures from a state health department and population figures from a federal source.
The source documentation was read first, and the indicator was defined as vaccinated children divided by the child population of the same county and year. The complication appeared in the trend: coverage in several counties jumped in one year, and the jump coincided with a switch in the population estimates used as the denominator and not with any change in practice. The series was rebuilt with a consistent set of population estimates, the year of the switch was marked as a break in the figure, and the counties were compared with differences in coverage and their confidence intervals instead of a simple ranking. The report listed the sources, versions and suppressed counties.
Outcome: The researcher received the source register, the indicator definitions, the coverage estimates with 95% confidence intervals, the trend figure with the break marked and the commands to rebuild the tables.
This example does not describe a real client or project. We never share details of actual client work, in line with the confidentiality we promise every researcher. This illustrative example is included to show the type, scope, and standard of work involved in this kind of analysis.
SOFTWARE
Software for Assembling and Comparing Public Data
Public files range from small summary tables to large microdata releases with survey weights, and each format suits some programs better than others.
HOW WE ASSESS YOUR PROJECT
We Review Your Actual Research
Your assessment is based on a review of your actual research materials and analytical requirements, not a generic project category.
You Submit Your Project Details
Share whatever is available at the current stage, such as the research aim, protocol or proposal, dataset, variable list, manuscript or statistical requirements.
We Review Your Research and Define the Scope
Your statistically trained Research Coordinator reviews the materials to understand the study design, research questions, data structure, analytical requirements and expected deliverables. If clarification is needed, additional documents or information can be requested before the assessment is finalized.
Your Analysis Is Completed and Delivered
Once the scope and fixed project fee are agreed, the in-house statistician completes the work and the Research Coordinator remains the communication point through delivery, with the analysis, deliverables and timeline as agreed.
RESEARCH RESPONSIBILITY
Policy Conclusions Rest With You
OnlineSPSS builds and checks the indicators within the agreed scope, while the policy question, the recommendations and the responsibility for the project stay with the researcher.
Agreed Statistical Support OnlineSPSS provides the agreed methodological, analytical and statistical reporting support, defined in a scope that is confirmed before any work begins.
No Intellectual Contribution OnlineSPSS does not claim any intellectual contribution to your work. Authorship of the capstone, thesis or report remains the researcher's.
Your Research Decisions The question the project asks, the choice of indicators and sources, the policy reading of the results and the conclusions remain yours, with overall responsibility.
Confidential Handling Research data and project details are handled confidentially, project information is not published, and an NDA is available where required.
FREQUENTLY ASKED QUESTIONS
Questions on Public Indicators and Their Sources
Indicators released only for larger areas, values hidden in public files and a check on whether a dataset suits the question.
What does public health surveillance data analysis include?
At OnlineSPSS it includes reading the source documentation, constructing the indicators, comparing places, groups or years with confidence intervals, and reporting the limits of each source. Sources range from surveillance reports and vital statistics to national surveys.
What if an indicator is published only for larger areas than my study needs?
Agencies often release an indicator only for states or large counties, both to protect privacy and because estimates for small populations are unstable. The options are to request a restricted-use file through the agency, to use model-based small-area estimates where an agency publishes them together with their uncertainty, or to set the question at the geographic level the data support. Dividing a larger area’s figure among its parts in proportion to population assumes that every part shares the same rate, which is rarely defensible.
What happens to suppressed or top-coded values in a public file?
Agencies hide small counts or cap extreme values to protect privacy, so those values cannot be recovered. A suppressed cell is not zero, and treating it as zero biases totals downward. The options are to combine categories or years until the counts can be shown, to report the cell as suppressed, or to use estimates only where the source supports them. Top-coded variables, such as the highest age or income group, are summarized with that cap in mind.
How can I tell whether a public dataset suits my question?
Check that it covers the population and years you need, measures your outcome with a definition close to your concept, includes the variables for the comparisons you plan, and releases enough detail for the groups of interest without suppression. The documentation also states how the data were collected and weighted. A dataset that fails one of these checks may still serve a narrower question, which should then be stated.
How do you protect my research data and confidentiality?
Your data and project details are handled confidentially, and project information is not published. Researcher names and institutions are not used publicly without permission. An NDA is available where required and can be arranged before sensitive materials are shared. Read more about confidentiality and data security.
What happens when I request an assessment, and how long does it take?
You share your research question, design and relevant materials, such as the names of the datasets, the indicators of interest and the files you have downloaded, and a statistically trained Research Coordinator reviews the request and responds the same day. The analysis, deliverables, timeline and one fixed project fee are then confirmed before any work begins, so the timeline follows from the agreed scope.
Still have a question? Find more answers in our full FAQ.
NEXT STEP
Indicators Defined
Before They Are Compared
A published figure looks final, but many carry definitions, suppressions and later revisions that any comparison across places or years has to respect.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014