EEOC Explore

EEOC Explore, User Guide (View & Build Dashboards)

Welcome! This guide is your friendly, start-to-finish companion for EEOC Explore, a set of easy-to-use dashboards built entirely from public data released by the U.S. Equal Employment Opportunity Commission (EEOC). Whether you simply want to look at the numbers, how many discrimination charges were filed, how lawsuits were resolved, how the trends are moving, or you want to build your own charts and dashboards, everything you need is in here.

Who this guide is for. Anyone on staff. You do not need to be technical, and you do not need to know anything about data, spreadsheets, or programming. If you can use a web browser, you can use EEOC Explore. We explain every button, label, and term in plain English, and we define any jargon the first time it comes up.

What you can do with EEOC Explore. Two things, really:

  1. View dashboards, open ready-made dashboards, narrow the numbers down to what you care about (a year, a state, a type of complaint), read exact values, and download the data or a picture of a chart.
  2. Build your own, use a point-and-click tool to make your own charts from public data, even ask an AI helper to build a chart from a plain-English description, and then download or share what you make.

How to use this guide. You don't have to read it cover to cover. If you're brand new, start at the top and work down, each section builds gently on the one before. If you already know your way around, use the Table of Contents below to jump straight to what you need. Steps are numbered so you can follow along click-by-click, and anything you can click in the app is shown in bold.

A reassuring note before we start: everything in EEOC Explore is built from public EEOC data, the same charge, litigation, and enforcement figures the agency releases to the public. There is no confidential or internal information behind the login. The dashboards are view-only, so you can click, filter, and explore freely without any risk of changing or breaking anything.

The site lives at one address: https://eeoc-dashboards-1-uecu.onrender.com. Open it in any web browser.

Table of Contents

  1. Signing in & getting around
  2. Reading a dashboard
  3. Ask the dashboard a question (AI search)
  4. Build your own dashboard (Explore tool)
  5. Downloading data & sharing
  6. Tips, troubleshooting & glossary
  7. Getting help

We'll begin where everyone begins: signing in and finding your way around the home page.

Signing in & getting around

This section walks you through logging in, what you'll see when you arrive, and how the dashboards are organized so you can find your way around. Everything here is built entirely from public EEOC data, the same charge, litigation, and enforcement figures the U.S. Equal Employment Opportunity Commission releases publicly, so there is no confidential or internal information behind the login.

There is one address for everyone: https://eeoc-dashboards-1-uecu.onrender.com. It is the everyday, view-only site. You reach it by typing the address into your web browser's address bar, the same place you'd type any website.

Signing in

If your copy of the app is set up to require a login, the first thing you'll see is a sign-in screen titled EEOC Dashboards, with the line "Authorized users only. Sign in to continue." underneath.

  1. Open your web browser and go to https://eeoc-dashboards-1-uecu.onrender.com.
  2. In the Email box, type your work email address. That is your username, there is no separate username to remember.
  3. In the Password box, type the password your administrator gave you. (You don't set this yourself; if you don't have one, ask your administrator.)
  4. The Authenticator code box is only for accounts that use two-factor authentication ("2FA", an extra one-time code from a phone app). It is labeled "(if your account uses 2FA)" and shows the hint "6-digit code (optional)." If your account doesn't use 2FA, leave this box empty.
  5. Click the blue Sign in button.

If the email or password is wrong, the page reloads and shows the message "Incorrect email or password." Just re-enter them and click Sign in again.

Depending on how your app is configured, you may instead see a standard browser pop-up asking for a username and password, or be sent to a separate government sign-in page. In all cases the rule is the same: your username is your email and the password comes from your administrator.

The "view-only" message you may see

Accounts come in two kinds: editors, who can change and build dashboards, and view-only users, who can open and explore everything but not change it. Most people are view-only, and that's all that's needed to read the dashboards.

If you're a view-only user and you land on an editing-only page, you'll see a screen headed View-only access that says: "Your account can view and explore dashboards, but editing and the builder are limited to editor accounts. Ask an administrator if you need edit access." This is normal and not an error. It simply means that page is reserved for editors. Click the ← Back to dashboards link on that screen to return to the main page, and carry on exploring as usual. If you genuinely need to edit, ask your administrator about an editor account.

The EEOC Explore landing page

Once you're signed in, you arrive at the home page, titled EEOC Explore, with the subtitle "Published Data Dashboards · U.S. Equal Employment Opportunity Commission." A round "EEOC" seal sits to the left of the title, and a thin bar at the very top reads "An official open-source conversion · U.S. Equal Employment Opportunity Commission data."

Below the title is a short introduction reminding you that these are "Published data dashboards, built entirely from public EEOC data sources." From here, everything is a click away on cards.

The middle of the page is a gallery, a grid of rectangular tiles called cards, each one a doorway into a dashboard. A card shows the dashboard's name in blue, a one-line description of what's inside, and a blue button to open it. To open any dashboard, click its button (for most cards this reads Open dashboard →). The cards rearrange themselves to fit your browser window, so on a narrow window they may stack into a single column.

The "Charge and Litigation Statistics" group and its sub-tabs

The main EEOC tool appears as a single card titled Charge and Litigation Statistics, described as "Charge Statistics, Litigation Statistics & Tables and Downloads, the full enforcement & litigation tool." Click its Open → button to go in.

Inside, you'll see a dark-blue banner reading Charge and Litigation Statistics with a row of tabs just beneath it. These tabs are the four sections of the tool, click any tab to switch between them:

  1. Charge Statistics, the charge-filing dashboard (this is the tab you land on first when you enter the tool).
  2. Litigation Statistics, the litigation dashboard.
  3. Tables and Downloads, the underlying figures laid out as tables, with options to download the data.
  4. Time Series & Forecasts, trends over time and projections built from the charge-receipts data.

The tab you're currently viewing is highlighted in dark blue; the others appear in light gray. Each of these four dashboards also has its own filters and, in some cases, its own inner tabs, but the four tabs above are always there at the top so you can jump between the sections without going back to the home page.

The "Build your own dashboard with public data" card

Near the top of the gallery is a special outlined card titled Build your own dashboard with public data. It describes the builder this way: "Pick or upload data, click columns onto X / measures / color, and the chart renders live. Every chart shows its editable Plotly code ('Open code'), or describe what you want and the AI writes it." Its button reads Open the explorer →.

This is the do-it-yourself tool for making your own charts from public data. Note that building and editing are limited to editor accounts, if you're a view-only user and click in, you'll see the "View-only access" message described above. (The builder itself is covered in its own section of this guide.)

The "← EEOC Explore" back link

Whenever you're inside a dashboard, look at the top-left corner for a link reading ← EEOC Explore. Click it at any time to return to the EEOC Explore home page and its gallery of cards. Think of it as the "home" button for the whole platform, no matter how deep you've drilled into a dashboard's tabs and filters, this link always brings you back to the start.

Now that you can find and open a dashboard, let's look at how to read one and narrow it down to the numbers you care about.

Reading a dashboard

Once you open a dashboard (for example, by clicking Open → on the Charge and Litigation Statistics card from the home page), you land on a page built from three parts: a row of tabs near the top, a column of filter cards down the left edge, and the charts and tables filling the rest of the space. This section explains how to move around, narrow the data down, and read what you see. Reading and filtering never change the underlying data, so you can explore freely. The dashboards are responsive: the layout reflows to fit your browser window, so charts and tables stay readable on a wide monitor or a narrow laptop screen.

Switching tabs

A dashboard groups related views onto tabs (a tab is a labeled page you click to switch between, like the tabs in a web browser). There are two levels of tabs:

  1. The blue EEOC Explore bar at the very top. Under the heading "Charge and Litigation Statistics" you'll see a row of tabs: Charge Statistics, Litigation Statistics, Tables and Downloads, and Time Series & Forecasts. These switch between the four major tools. The tab you're currently on is filled solid blue; the others are pale gray. Click any one to jump to that tool.
  2. The dashboard's own tabs, lower down. Just above the charts, each tool has its own row of tabs for its sub-views (for example, Charge Statistics has several). Click a tab label to show that view. The charts and filters below refresh to match.

To change what you're looking at, click the tab you want. Nothing is "submitted" or saved, switching tabs only changes the display.

Filters and parameters: re-slicing every chart

Down the left side of the dashboard is a column of small white filter cards. Each card has a bold title (for example, Fiscal Year, Discrimination Type, or Sex) and a control underneath it. These let you narrow the data, and importantly, every chart and table on the page redraws together to reflect your choices. There is no "Apply" button; the page updates the moment you make a selection.

The cards come in a few shapes:

Above the tabs you may also see one or more parameter dropdowns (a parameter is a workbook-wide setting, shown with its own caption to the left). These behave like a filter card but often control something structural, for instance, which data table or which measure the charts display. Changing a parameter also redraws everything below it.

A few things to know:

Reading the charts

The main area holds the charts themselves, bar charts, line charts, US maps (choropleths, where states or counties are shaded by value), point maps, and big single-number tiles (a headline figure such as a total count). Each chart usually has a centered title above it describing what it shows; titles often update to reflect your filters (for example, the year you selected).

You read these the ordinary way: taller bars and darker map shading mean larger values; a line chart's slope shows a trend over the years on its horizontal axis. Where a chart breaks data out by category (race, sex, basis), a legend identifies each color.

Hovering for exact values

Charts show the shape of the data; to read an exact number, move your mouse pointer over a bar, a line point, a map area, or a plotted dot and pause. A small tooltip pops up next to the cursor with the precise value (and, on maps, the state or area name). This is the reliable way to get the underlying figure rather than eyeballing it against the axis. Move the pointer away and the tooltip disappears. (On a touch screen, tap and hold the element instead.)

Reading the data tables

Some tiles are data tables rather than charts, rows and columns of values, with a shaded header row. A few ways to work with them:

The dedicated Tables and Downloads tab works the same way but is built for full crosstabs: pick a table from Select a Data Table and optionally one or more years under Fiscal Year, and the table redraws. Its header row stays pinned as you scroll.

The "More info" column and → links

In tables that list litigation resolutions, you'll see a narrow More info column. Instead of showing a long web address, each row holds a single → (arrow). The arrow is a link: click it to open the EEOC press release for that case (the official write-up of how the lawsuit was resolved).

The press release opens in a new browser tab, so your dashboard, including all the filters you've set, stays exactly as you left it on the original tab. When you're done reading, close the new tab (or switch back to the dashboard tab) to continue. If a row has no associated press release, its More info cell is simply blank.

Downloading a single chart as a PNG image

On the dashboards that support it, the Time Series & Forecasts chart and any chart on a Build your own dashboard page, you can save a chart as a PNG image (a standard picture file you can drop into a slide deck, email, or report). To do this:

  1. Move your mouse over the chart. A small toolbar of icons appears in the chart's top-right corner.
  2. Click the camera icon (its tooltip reads "Download plot as a png").
  3. Your browser downloads a PNG of that single chart, rendered at high resolution. Look in your usual Downloads folder.

This saves only that one chart, not the whole dashboard or its data. (If you instead want the numbers behind the dashboard, use the ⬇ Download data (CSV) button in the dashboard's header, which gives you the underlying tables as spreadsheet files, see "Downloading data & sharing" below.)

Note: on the main Charge Statistics and Litigation Statistics dashboard tiles, this hover toolbar is intentionally turned off to keep the layout clean, so the camera icon does not appear there, use the dashboards listed above, or the CSV download, when you need to export.

Reading a dashboard is one way to get answers. Often the fastest way is simply to ask, in plain English.

Every dashboard has a built-in search box that lets you ask a question in plain English, like talking to a person, and get a written answer, a chart, and a table back. You don't need to know anything about the data structure or write any formulas. Behind the scenes an AI model (a computer program that understands ordinary language) reads your question, looks up the numbers in this dashboard's own data, and writes you a short answer.

You'll find this on every dashboard page, just under the dashboard's title and the toolbar buttons, and above the filters and chart tabs.

Asking a question

  1. Find the long text box near the top of the dashboard. It shows greyed-out hint text that reads: Ask this dashboard a question, e.g. "how many charges were filed in 2024?"
  2. Click in the box and type your question in normal English.
  3. Press Enter, or click the blue Ask button to the right of the box.
  4. A small animated dot appears while the AI works. An answer usually takes a few seconds; a harder question can take longer. Wait for it to finish, you don't need to click anything again.

You can ask another question at any time by clearing the box, typing a new one, and pressing Ask again. The new answer replaces the old one.

Questions that work well

The AI answers best when your question is about counts, totals, rankings, or trends in this dashboard's numbers. Good examples:

Ask about one thing at a time, and name the years, states, or categories you care about. Each dashboard only knows about its own subject, so ask the charge-data dashboard about charges, the EEO-1 dashboard about private-employer headcounts, and so on.

What comes back

When the AI answers, a white card with a dark blue left edge appears just below the search box. Depending on your question, it can contain up to three things, in this order:

  1. A plain-English summary, one or two sentences at the top, in bold, stating the answer (for example, the total or the leading state). This is the quick takeaway.
  2. An auto-generated chart, when the answer is something that can be drawn, the platform builds a chart for you automatically: - a line chart when the answer is a trend across years (you'll see a point for each year connected by a line), or - a bar chart when the answer is a ranking of categories such as states or industries (showing up to the top 20, tallest bar first).

Not every answer produces a chart. A single number (like one grand total) has nothing to plot, so the card simply shows the summary and table instead. The chart is a normal interactive chart, you can hover over it to read exact values. 3. A data table, the underlying rows behind the answer, with column headings, showing up to the first 10 rows. This is the raw detail the summary and chart were built from.

Check the numbers before you trust them

Underneath the answer is a small grey line you can click: Show the query it ran (verify the numbers). Clicking it expands a box that shows the exact database query (the precise instructions the AI wrote to pull the numbers) in plain text.

Why you should look: the AI writes these instructions itself, and it can occasionally misread your question, for example, leaving out a filter, counting the wrong column, or including a category you didn't mean. The summary will still sound confident even when the query was slightly off. Expanding this box lets you (or a colleague) confirm the query actually asked for what you wanted before you put the number in a report, email, or briefing. If the query doesn't match your intent, rephrase your question more specifically and ask again. Treat the AI's answer as a fast first draft to be verified, not as an official figure.

"Answered by …", which AI answered, and how long it took

At the very bottom of the answer card, in small grey text, is a note such as Answered by the local model qwen2.5-coder:7b in 4.2s from this dashboard's data, verify the query above for anything you report.

This tells you two useful things:

If you see a yellow warning instead of an answer

If the AI can't answer, a yellow box appears in place of the answer card with a short explanation. The most common cause is that no AI model is running on the machine, in that case the message will tell you to start the local AI service. If you see this, let whoever set up the platform know; ordinary view-only use of the site doesn't let you start it yourself.

So far you've been working with ready-made dashboards. Next, let's build your own from scratch.

Build your own dashboard (Explore tool)

The Explore tool (the "Chart Explorer") is a point-and-click chart builder. You pick a data set, choose a chart type, and drag your columns into roles like X-axis and measurement, the chart redraws instantly every time you change something. Behind every chart is real, editable code, and there's an AI helper that can build a chart from a plain-English description. This section walks through it step by step.

A quick note on who can do what: any signed-in user can build charts and download them. Only editor accounts can save a dashboard into the shared, published catalog that everyone sees. If you're not an editor, you can still do everything described here, you just download or share each chart as a standalone file instead of saving it to the catalog. This is called out again where it matters below. (One thing to note: some setups also limit opening the builder to editor accounts, if you're a view-only user and the builder won't open, you'll see the "View-only access" message from the first section, and you should ask an administrator for an editor account.)

Opening the explorer

  1. From the home page, click Build your own dashboard with public data.
  2. The Chart Explorer opens. You'll see a dark blue title bar reading Chart Explorer with the subtitle "Slice · dice · live Plotly code · AI", and a ← Studio link in the top-right corner that takes you back to the home page at any time.
  3. The screen is split into two columns: a narrow control panel on the left (data, chart settings, slice-and-dice, and AI) and a large center area showing the chart, its code, and dashboard controls.

Picking a data source

At the top of the left panel is the Data source section. You have two ways to get data in:

  1. Click the Pick a dataset… dropdown and choose one of the available public EEOC data sets. (These are the converted data sets already on your machine.)
  2. Or upload your own file: click the ⬆ or upload a file box and select a file from your computer. Once it loads, the explorer saves it as a new data set you can chart.

After a data set loads, a green confirmation line appears, for example ✓ <name>: 12,000 rows × 8 cols, telling you how many rows and columns it found. If something goes wrong, you'll instead see a "Could not load…" message explaining why.

To save you a step, the explorer picks sensible starting fields automatically, it puts the first text column on the X-axis and the first number column as the measure, so a chart usually appears right away.

Choosing a chart type

In the Chart section, use the Chart type dropdown to pick how your data is drawn. The full list of chart types is:

Below the chart type are the field controls. Set them to match the chart you want:

  1. X / dimension, the category or axis along the bottom (for a map, this is the column holding state codes).
  2. Measures (Y), the number(s) being plotted. You can pick more than one. For a Bubble chart, the second measure you pick controls the bubble size.
  3. Color / split, optionally break the data into colored series by a column.
  4. Aggregate, how repeated values are combined: sum, mean, median, min, max, count, or none (don't combine). The default is sum.

If a chart needs a field you haven't filled in, you'll see a gentle prompt in red, like "⚠ pick the fields this chart needs…", fill in the missing field and it renders.

Slice and dice

The Slice & dice section lets you narrow and reshape the data. Every change here re-renders the chart instantly, there's no "apply" button.

  1. Filter column, choose a column you want to filter on.
  2. Keep values, once you've picked a filter column, this dropdown fills with that column's actual values. Select one or more to keep only those rows (leave it on "all" to keep everything).
  3. Sort, choose No sort, High → low, or Low → high to order the results by the measure.
  4. Top N, type a number to keep only the top (or bottom, depending on sort) N results. Leave it blank for all.
  5. Theme, restyle the chart's look. Options are plotly_white (the default), plotly, plotly_dark, ggplot2, seaborn, simple_white, and presentation.
  6. Animate by (play ▶), optionally pick a column to animate the chart across; this adds a Play button so the chart steps through that column's values (for example, year by year).

The "Open code" panel, peek at or edit the chart code

Every chart is backed by real Plotly/Python code, and you can see and change it. (Plotly is the charting library that draws every chart in this app; Python is the programming language it's written in, but you don't need to know either to use the controls above.) Below the chart is a row labeled ◇ Open code (editable Plotly/Python), and underneath it a text box showing the exact code that drew the current chart.

Note on safety: this code runs locally on your own machine in a restricted sandbox. A few things are deliberately blocked (importing libraries, reading files, system access), the tools you need (df, pd, np, px, go) are already provided for you. Your code just needs to end by assigning the chart to a variable named fig.

AI: build a chart, or get insights

The AI section (near the bottom of the left panel) has two helpers. Both need a local AI model running, if none is available, you'll see a message such as "No AI model, start Ollama or set OPENAI_API_KEY."

Build a chart from a description:

  1. In the text box, type what you want in plain English, for example, "charges by state as a map." (The box shows this as a hint.)
  2. Click ✨ Build.
  3. The AI writes the code, drops it into the Open code box, and draws the chart. You'll see "✓ Built, tweak the code or controls." From here you can refine it with the controls or by editing the code and clicking ▶ Run. If the AI's code doesn't run cleanly, you'll be told to edit it and Run.

Get insights about the data:

  1. Click 💡 Insights.
  2. The AI reads a summary of the data set and returns 3-5 short, factual bullet points, notable totals, leaders, gaps, or trends, written for a non-technical reader. These appear in a light blue box under the buttons.

Downloading a chart (the camera icon)

You don't need any special permission to download a chart, anyone can do this.

  1. Hover your mouse over the chart. A small toolbar appears in the chart's top-right corner.
  2. Click the camera icon ("Download plot as a png").
  3. The chart saves to your computer as a PNG image (saved at 2× scale for a crisp, high-resolution picture).

This is the way every user, editor or not, gets a copy of a chart to drop into a document, email, or slide deck.

Building a multi-chart dashboard, and saving vs. sharing

You can group several charts into one dashboard.

  1. With a chart you like on screen, click ➕ Add to dashboard (in the row next to ▶ Run). The chart is validated and added to a running list, shown under the 📊 Dashboard heading. You can remove any chart from the list with its ✕ button.
  2. Give the dashboard a name in the Dashboard name box (it starts as "My Dashboard").
  3. Click 💾 Save dashboard.

What happens next depends on your account:

Once you have a dashboard you like, ready-made or your own, here's how to get the data out of it and share it with others.

Downloading data & sharing

Every dashboard has two buttons in its top bar, just to the right of the dashboard's title: ⬇ Download data (CSV) and 📤 Publish / Share. You do not need editing rights to use either one. This section explains what each one does and walks you through using them.

Getting the data behind a dashboard (CSV download)

If you want the raw numbers, for example to open them in Excel, double-check a figure, or feed them into your own report, use the data download. CSV ("comma-separated values") is a plain spreadsheet file that Excel, Google Sheets, and Numbers all open directly.

  1. Open the dashboard you want.
  2. In the top bar, next to the dashboard's title, click ⬇ Download data (CSV).
  3. Your browser downloads a single ZIP file (a compressed folder) named after the dashboard, ending in _data.csv.zip, for example Charge_Statistics_data.csv.zip.
  4. Find the file in your Downloads folder and double-click it to unzip it. On Windows you may need to right-click and choose Extract All.
  5. Inside you'll find one CSV file for each underlying data table. Each file is named after its table (for example charges.csv). Open any of them in your spreadsheet program.

This gives you all of the dashboard's underlying data, every table, not just the slice currently shown on screen. (It's the same idea as Tableau's "Download to Data.") The same ⬇ Download data (CSV) button also appears on the dashboard's Tables and Downloads page if you'd rather grab it from there.

Browse the whole data catalog (one page for everything)

Besides the data behind any one dashboard, the site has a single page that lists all of the published public data in one place. To see it, go to /catalog (add /catalog to the end of the site address, so https://eeoc-dashboards-1-uecu.onrender.com/catalog).

The catalog page shows every dataset, the tables inside it, how many rows each one has, and the plain-English meaning of each column (which columns are categories you group by, and which are numbers you measure). It is a read-only page; you do not need to be an editor to open it, and nothing on it can be changed.

At the top of the catalog page is a Download data catalog and inventory link. Click it to download a single ZIP file (a compressed folder) holding the full inventory: a machine-readable catalog, a description of each dataset, the column dictionaries, the inventory as both a JSON and a spreadsheet (CSV) file, and a short README that explains what is inside. This is the right download when you want a complete, self-describing copy of what data the site publishes, for a records request, an audit, or your own reference.

Sharing a dashboard

Clicking 📤 Publish / Share in the dashboard's top bar opens a Publish & Share page. The top bar of that page has two links on the right, ↗ open (jumps to the live dashboard) and ← all dashboards (returns to the home list). Below that are three cards, each a different way to share. The first two are described here.

The first card, 🔗 Share a live link, gives coworkers a web address that opens the live, interactive dashboard in their own browser, filters, search, and all. The link points at the site, https://eeoc-dashboards-1-uecu.onrender.com, so anyone you send it to can open it from any browser.

The card shows a box with a Copy button holding the dashboard's link.

To share it:

  1. On the Publish & Share page, find the 🔗 Share a live link card.
  2. Click Copy. The button briefly changes to Copied! to confirm.
  3. Paste the link into an email, chat message, or calendar invite for your coworker.
  4. When they open the link, they see the live dashboard.

Note: anyone who opens this link gets a view-only dashboard. They can explore and filter, but they cannot edit it or change anything you've built.

Option 2: A single self-contained HTML file you can email or post

The second card, 📄 Standalone HTML file (no server needed), builds one self-contained file containing the dashboard. "Self-contained" means everything needed is baked into that one file, the person who opens it does not need the app running, does not need to be on your network, and does not need anything installed. They just double-click it and it opens in any web browser. This is the right choice for emailing a dashboard to someone outside the office or posting it on a shared drive.

One thing to know: the charts in the file are still interactive (you can hover over them for details), but the live features, the all-charts filter and the AI "Ask" search box, only work in the running app, so they're left out of the standalone file. It captures the dashboard's default view (for example, the most recent fiscal year).

The card has two buttons:

To send someone the file:

  1. On the Publish & Share page, find the 📄 Standalone HTML file (no server needed) card.
  2. (Optional but recommended) Click View standalone first to check it looks right.
  3. Click ⬇ Download .html and note where it saves (usually your Downloads folder).
  4. Attach that .html file to an email, or drop it onto a shared network drive or SharePoint.
  5. The recipient saves the file and double-clicks it; it opens in their browser with no further setup.

(The third card on this page, 📦 Portable app bundle (.zip), packages the dashboard as a runnable folder for a developer to host elsewhere, it's a technical hand-off, not something a non-technical reader needs.)

Finally, here are quick fixes for common snags and a plain-English glossary of the EEOC terms you'll meet on the charts.

Tips, troubleshooting & glossary

This section covers the small problems you are most likely to run into while using the EEOC Explore dashboards, and a plain-English glossary of the EEOC terms you will see on the charts. The site everyone uses is at https://eeoc-dashboards-1-uecu.onrender.com and is view-only.

Common issues and what to do

"Incorrect email or password" when you try to sign in

When you open the site, you may see a blue Sign in screen titled "EEOC Dashboards" with three boxes: Email, Password, and Authenticator code (the last one is only needed if your account uses two-step sign-in). If you enter your details and get a red "Incorrect email or password" message, work through these steps:

  1. Check that your email and password are typed exactly right, passwords are case-sensitive, so watch for Caps Lock and any extra spaces.
  2. If your account was set up with two-step sign-in (also called 2FA, short for "two-factor authentication"), the Authenticator code box is required. Open your authenticator app (such as Google Authenticator or Authy), read the current 6-digit code, and type it in before clicking Sign in. If you see "Wrong or missing authenticator code," the code has expired, wait for the app to show a fresh one and try again.
  3. If it still won't let you in, your email may not be on the approved list, or your password may need to be reset. Only people who have been added by an administrator can sign in, so contact your administrator to confirm your email and password.

The AI search is slow to answer

Every dashboard has an Ask box near the top where you can type a plain-English question (for example, "how many charges were filed in 2024?") and click the Ask button. The answer is written by an AI model. Here is what to expect:

  1. The site tries a fast model running on your own machine first. If that local model is busy or takes too long, the site automatically switches to a cloud-based model to finish the job, so an answer that takes a few extra seconds is normal. Just wait a moment.
  2. When the answer appears, a small gray line underneath tells you which model produced it, for example "Answered by the local model qwen2.5-coder:7b" or "Answered by OpenAI gpt-4o-mini." That line is just telling you where the answer came from; you do not need to do anything with it.
  3. If you see a yellow warning box saying no AI model is available, the local AI helper (Ollama, the program that runs the on-machine AI model) isn't running and no cloud model is set up. That is a setup issue for whoever runs the server, not something you can fix from the dashboard, let your administrator know.
  4. Whenever you plan to quote a number from an AI answer, click "Show the query it ran (verify the numbers)" underneath the answer to see exactly how it was calculated, and double-check it against the charts.

You see a "View-only access" message

If you click an editing tool and get a page headed "View-only access" that says editing and the builder are limited to editor accounts, this is expected. It is not an error. The site is deliberately read-only for everyone, so the published dashboards can't be changed by accident. You can still open every dashboard, use the filters, ask questions, and download the data. If your job requires you to edit or build dashboards, ask an administrator for an editor account. To get back, click "← Back to dashboards."

Glossary of EEOC terms

These are the terms you will most often meet on the charge and litigation dashboards. Each is given in one plain sentence.

Getting help

If you get stuck, here's where to turn:

When you reach out for help, it speeds things up to mention the exact wording of any message you saw and what you were trying to do at the time.

↑ Back to top