> ## Documentation Index > Fetch the complete documentation index at: https://help.voxdash.com/llms.txt > Use this file to discover all available pages before exploring further. # Results and Export > Review every call and sample record of an AI-CATI study, read the AAPOR outcome and response-rate tables, reconcile cost, and export metadata, raw data, transcripts and recordings. The **Results** tab shows what the study has produced. It opens once the study has started dialling (for a study with sub-samples, once the **main launch** has dialled; each sub-sample run has its own results on its run page). Pre-test results are on each pre-test's own page. ## Calls and records The table at the top of the tab has two views: ![Results toolbar in the Per call view](https://mintcdn.com/voxdash/OFPcyQtvMysHCjUR/images/AICATI/c-results-toolbar-per-call.png?fit=max\&auto=format\&n=OFPcyQtvMysHCjUR\&q=85\&s=fe39ee0a1185ea920614a1997d8289ef) ### Per call Every dial is listed as soon as it is placed. | Column | Description | | - | - | | **Case** | The record's case ID. | | **Number** | The phone number, masked unless **Show phones** is on. | | **Call time** / **Duration** | When the call was placed and how long it lasted. | | **Status** | The call's outcome (its AAPOR disposition). Filter by it with **All statuses**. | | **Quality** | STT confidence, reply latency in milliseconds, and **Speedster** or **Straight-lined** flags. | | **Settings** | The settings version the call ran under (see [Settings versions](#settings-versions)). | | **Open** | Opens the [sample record](https://help.voxdash.com/AI-CATI/Sample-Records.md). A headset icon marks calls with a recording. | Filters: search by case or number, call status, a **Calls from / Calls until** date range, and **Recorded only**. ### Per sample record ![Results toolbar in the Per sample record view](https://mintcdn.com/voxdash/OFPcyQtvMysHCjUR/images/AICATI/c-results-toolbar-per-record.png?fit=max\&auto=format\&n=OFPcyQtvMysHCjUR\&q=85\&s=718f442ca0c46a06e8f3eb78e19d3a00) Each record appears with its latest outcome once it has been dialled. | Column | Description | | - | - | | **Case**, **Number** | Case ID and (masked) phone number. | | **Region ยท TZ** | The record's region and time zone. | | **Call count** / **Last call** | How many calls were placed and when the last one was. | | **Status** / **State** | The latest outcome, or the record's own state when it has none yet: **Pending** (not called yet), **In progress**, **Resolved**, **Unused**. | | **Next attempt** | When it will be called again. An icon marks an **appointment booked with the respondent**. | | **Open** | Opens the sample record. | Filters: search, **All time zones**, **All states**, and **Include records not called**. > [!TIP] > > Filters are kept in the page address, so **Back to results** on a record page returns you to the same filtered view. > [!WARNING] > > **Show phones** reveals full phone numbers on screen. Only turn it on when you need it, and never share screenshots with numbers visible. ## Analysis panels Below the table, the Results tab shows the study's reporting panels. Each has a **Download** button for its figures. ### AAPOR outcome table (worked sample) Every worked record placed in an AAPOR outcome bucket, with **Code**, **Label**, **Count** and **Share of worked**, plus the total. The **unknown-eligibility model** shows how records with unknown eligibility are treated when the rates are worked out; it comes from the coding standard shown on the Setup tab. If some records could not be placed in a bucket, their number is shown. ### Response, cooperation and contact All rates follow the AAPOR standard definitions. **RR3** is the headline response rate. | Rate | Definition | | - | - | | **RR1** | Completes divided by all eligible cases plus every case of unknown eligibility. The most conservative response rate. | | **RR2** | Like RR1, but partial interviews count as responses too. | | **RR3** (headline) | Completes divided by eligible cases plus only the estimated eligible share of unknown-eligibility cases. | | **RR4** | Like RR3, with partials counted as responses. | | **RR5** | Completes divided by known eligible cases only. | | **RR6** | Like RR5, with partials counted. The most generous response rate. | | **COOP1** | Completes among all eligible people reached, including those who could not take part. | | **COOP2** | Like COOP1, with partials counted as cooperating. | | **COOP3** | Completes among eligible people reached, leaving out those unable to take part (for example language or illness). | | **COOP4** | Like COOP3, with partials counted. | | **CON1 / CON2 / CON3** | Share of cases in which someone was reached, counting every / the estimated eligible share of / none of the unknown-eligibility cases as eligible. | | **REF1 / REF2 / REF3** | Refusals and breakoffs as a share of eligible cases, with the same three treatments of unknown eligibility. | | **e-factor** | The estimated share of unknown-eligibility records that are in fact eligible. Some rates use it. | | **Operational cooperation** | The cooperation the calling engine achieved on the calls it actually placed. Not a standard AAPOR rate. | ### Cost reconciliation | Figure | Meaning | | - | - | | **Estimated** | What the estimate in step 7 priced. | | **Cost so far** / **Final cost** | What the study has spent. | | **Spend cap** | The maximum authorised spend. | | **Cost of goods sold** | What was paid to providers (telephony, speech, AI model) for these calls, before plan markup. | | **Variance** | Spend so far minus what the estimate priced for the same calls. Above zero means the calls cost more than estimated. | | **Per complete** | Cost per completed interview. | **Cost by disposition** splits the cost by call outcome (Outcome, Calls, Cost, Share). ### Sample accounting Cleaned sample, worked sample, **sample coverage** (worked over cleaned), **yield** (completes over worked), records in progress, records **used in pre-test (excluded)** and **unused** records. ### Achieved versus planned Achieved against the distribution targets set in [step 5](https://help.voxdash.com/AI-CATI/Setup-Screening-Targets.md). A deviation of **10 percent or more** is highlighted. ### Operational, cost and technical-health reports Downloadable reports such as the **full per-call log**, calls log, dispositions, response rates, AAPOR outcomes, cost reconciliation, cost by disposition, sample accounting and targets by stratum. A file with more rows than the download limit cannot be downloaded from here; use the export instead. ### Settings versions Settings are sealed when the campaign launches. If you edit settings while the study is stopped and then resume, a new **settings version** is adopted. A notice on the Results tab tells you which version was in force for which calls, and the **Settings** column on each call shows it. The full list of versions, who adopted them and their effects is in **Settings history** on the Setup tab. ## Export data Select **Export** in the toolbar to export the entire study. ![Export data dialog with Metadata only selected](https://mintcdn.com/voxdash/OFPcyQtvMysHCjUR/images/AICATI/c-export-data-metadata.png?fit=max\&auto=format\&n=OFPcyQtvMysHCjUR\&q=85\&s=2c3de4cdb6b4f6fee175cc77ddd5fdb5) **1. Choose what to export** * **Metadata only:** case IDs, dispositions, paradata, timestamps. No response values or recordings. * **Raw data only:** response variables and coded values. No paradata or recordings. * **Everything:** metadata, raw data, transcripts and recordings. **2. Choose the format** * Metadata and raw data: **SPSS (.sav and .sps)** or **CSV**. SAS and Stata are listed but not available yet. * Everything: **ZIP archive**. **3. Choose which run** **Production run**, or one sub-sample run that has data. Pre-test runs are exported from their own page. **4. Generate or request the export** Metadata and raw data are generated and downloaded. An **Everything** archive is built in the background and **emailed to you** when it is ready. ![Export data dialog with Everything selected](https://mintcdn.com/voxdash/OFPcyQtvMysHCjUR/images/AICATI/c-export-data-everything.png?fit=max\&auto=format\&n=OFPcyQtvMysHCjUR\&q=85\&s=f245373f058b612b0f5b0fe92486d530) **Recent exports** lists earlier exports with their status: **Preparing**, **Ready** (its files are listed, ready to download) or **Failed** (with the reason, if any). Only one archive can be prepared for a study at a time. > [!NOTE] > > This export covers the whole study. To export one record, use **Export record** on its [sample record page](https://help.voxdash.com/AI-CATI/Sample-Records.md). | Message | What it means | | - | - | | No interview has been saved for this run yet | There is no data to export for the chosen run. | | The questionnaire this study is bound to no longer matches the version it was bound with | The file was not exported, because its columns might not describe the questions that were asked. | | Recordings storage is not available for exports here | The archive with recordings could not be written. | | This export stopped responding and was abandoned | Request it again. | ## Related pages * [Live Monitoring](https://help.voxdash.com/AI-CATI/Live-Monitoring.md) * [Sample Records](https://help.voxdash.com/AI-CATI/Sample-Records.md) * [Sub-samples](https://help.voxdash.com/AI-CATI/Sub-samples.md) This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.