VxAdmin supports a variety of results exports.

Tally reports contain contest results for the full election or a specific subset of ballots. They can be printed directly from VxAdmin, exported as PDFs, or exported in CSV format. In addition, full election results can be exported as an Election Results Reporting CDF JSON file.

Ballot count reports include only ballot counts, without contest results. They can also be printed directly from VxAdmin, exported as PDFs, or exported in CSV format.

The write-in adjudication report acts as a summary of all write-in adjudication activity or as a "scatter report." The tally reports consolidates write-in candidates whose vote totals are too small to affect the contest outcomes. The counts for those write-in candidates are available in the write-in adjudication report. The write-in adjudication report may be printed directly from VxAdmin or exported as a PDF.

For all VxAdmin reports, the report is either "Official" or "Unofficial." If the report is exported before the results are marked as official in the application, the results are "Unofficial." If the report is exported after the results are marked as official in the application, the results are "Official." All visual reports have "Official" or "Unofficial" included in the title and all exported files will have "official" or "unofficial" included in the filename.

The full election tally reports can be exported as JSON Election Results Reporting CDF results. The feature is only available for the full election results and not grouped or filtered results. The export follows the CDF specification without extensions. The CDF fields are used as follows:

The Election Results Reporting CDF requires that identifiers be in NCName ID format, so all identifiers from the election definition are re-formatted to be compliant. For example, "Sample County" becomes "vx_sample-county".

Basic Structure

The basic structure of the tally reports printed from VxAdmin is similar to the polls opened and closed reports printed from VxScan.

Filtered and Grouped Reports

In addition to the full election report, VxAdmin exports reports filtered or grouped by the following dimensions:

• Ballot Style

• Precinct

• Voting Method

• Scanner

• Batch

• District (filtering only)

When a report is grouped, the report is broken up into multiple sections each with their own filter. For example, if you were to group by "Voting Method," the resulting report would have a section filtered for precinct ballots only and a section filtered for absentee ballots only:

The filter for a given report section is indicated in the title of the report. If the report has more than one filter applied to it, the full details of the filter will be specified in a box below the title:

Note that filters and groups can be combined.

Manual Results

If manual results have been added, they will be included in the tally report alongside the scanned results. For each contest, there will be three results columns: "scanned", "manual", and "total".

Write-In Candidate Aggregation

In order to prevent a long list of write-in candidates from making contest results difficult to read, VxAdmin tally reports consolidate write-in candidates if they're not relevant to the contest outcome. Relevant write-ins are included as "<Name> (Write-In)". Irrelevant write-ins are consolidated as "Write-In" or, if there are relevant write-ins in the list, "Other Write-In". Unadjudicated write-ins are always consolidated as "Unadjudicated Write-In."

A write-in is deemed relevant if it must be listed for the reader to determine the winner(s) of the contest. In the middle example below, the single adjudicated write-in is consolidated because Natalie Portman is the winner regardless of the identities of the write-ins. In the last example, however, there are 31 adjudicated write-ins so it's possible that a write-in got more votes than Natalie Portman. The report then enumerates the write-ins with the highest vote totals until the "Other Write-In" count is smaller than the counts for all winners.

The write-in adjudication report includes the results of all write-in adjudication organized by contest:

The values in the contest table above are as follows:

Relationship to Tally Reports

Before adjudication, write-ins are treated as votes for a generic write-in candidate and will appear in tally reports as "Unadjudicated Write-In." After adjudication, the new value will be reflected on tally reports:

• When adjudicated for an official candidate, the write-in will be added to the official candidate count just as a regular vote for the candidate would be

• When adjudicated for a write-in candidate, either:

  • The write-in will appear in the adjudicated write-in bucket if it is not relevant to the results

• When invalid, the write-in becomes an undervote

Unmarked Write-Ins

Unmarked write-ins behave differently in relationship to the tally results because they cannot be considered a valid vote before adjudication. As a result, they are considered undervotes before adjudication. If an unmarked write-in is adjudicated as invalid, then tally results will not change. If an unmarked write-in is adjudicated for a candidate, the unmarked write-in is no longer an undervote and is added to the vote total for that candidate.

Also note that, because the write-in adjudication report includes unmarked write-ins, the "Unadjudicated Write-In" count on tally reports may be less than the "Not Adjudicated" count on the write-in adjudication report. The initial difference is the count of unmarked write-ins.

Ballot count reports contain ballot and ballot sheet counts, not contest results.

The ballot counts are broken down by ballot type - "HMPB" (for hand marked ballots), "BMD" (for machine marked ballots), and "manual" (for manually entered results, if applicable).

If there are multi-sheet ballots, the hand marked ballot counts may have sheet counts specified. The ballot count is determined by counting the first sheet of each ballot style.

When grouping is applied in a ballot count report, each group corresponds to a row in the table. In the following example, the ballot counts are grouped by both precinct and voting method:

The following metadata columns may appear in a ballot count report table in order to specify groupings:

Filters apply to the entire report. Simple filters are shown in the title, while complex filters are listed in a box below the title.

Both tally reports and ballot count reports can be exported in comma-separated values (CSV) format.

Tally Report CSV Structure

In the CSV tally report, the vote total for each candidate or contest option is listed in a single row. In addition, there are rows for overvote and undervote totals for a contest.

For example, here is an excerpt of a tally report including the results for one contest:

Precinct,Precinct ID,Contest,Contest ID,Selection,Selection ID,Total Votes
West Lincoln,20,Mayor,mayor,Sherlock Holmes,sherlock-holmes,0
West Lincoln,20,Mayor,mayor,Thomas Edison,thomas-edison,0
West Lincoln,20,Mayor,mayor,Write-In,write-in,0
West Lincoln,20,Mayor,mayor,Overvotes,overvotes,0
West Lincoln,20,Mayor,mayor,Undervotes,undervotes,0

The "Precinct" and "Precinct ID" columns are metadata fields that are included because this example export groups results by precinct. The other fields are standard fields:

If manual results were entered, two additional columns will be added - "Scanned Votes" and "Manual Votes." These columns denote which votes for each selection came from the scanners vs manual entry.

The results of write-in adjudication are always included in the CSV exports. All write-in candidates will appear with their adjudicated name and a UUID assigned by VxAdmin. Unadjudicated write-in candidates will appear as "Unadjudicated Write-In" with the ID "write-in". Unlike in the printed reports, write-in candidates are never consolidated.

Ballot Count Report CSV Structure

The ballot count report CSV is the same as the table presented in the printed ballot count reports. For example:

Precinct,Precinct ID,Voting Method,BMD,HMPB,Total
West Lincoln,20,Precinct,0,0,0
West Lincoln,20,Absentee,0,0,0
East Lincoln,21,Precinct,0,0,0
East Lincoln,21,Absentee,0,0,0
South Lincoln,22,Precinct,0,0,0
South Lincoln,22,Absentee,0,0,0
North Lincoln,23,Precinct,0,0,0
North Lincoln,23,Absentee,0,0,0

The "Precinct", "Precinct ID" and "Voting Method" columns are metadata fields that are included because this example export groups results by precinct and voting method. The other fields are standard fields:

Shared Metadata Structure

Any filters or groupings which apply to each row will be indicated in the row itself by metadata columns. For example, in the example export above, the "Precinct" column lists which precinct group each row represents.

If a row is filtered by a single attribute value (e.g. represents one precinct rather than multiple precincts) then the following basic metadata fields are used:

If a row is filtered by multiple attribute values, the following columns may be used:

In cases as above, where a row value includes comma-separated values, those values will be wrapped in quotation marks per typical CSV formatting in order to allow consumers to properly differentiate columns.