---
title: "Advanced crosstab analysis"
url: "https://helpdesk.telmarhelixa.com/en/article/advanced-crosstab-analysis"
language: "en"
last_updated: "2026-10-27T15:15:53.166Z"
---

# Advanced crosstab analysis

Covers N-Tiles, Volumetrics, Add to Multiple and Large Row Analysis. Custom Audiences and Snapshots are in Explore - Custom Audiences and Snapshots.

## **N-Tiles**

N-Tiles (quantiles) split continuous distributions (spend, age, satisfaction and so on) into equal intervals in ascending order. They are used for segmentation and benchmarking.

**Uses**

-   **Segmenting responses:** see how low, medium and high values on a metric differ in attitudes or behaviors. Example: Quartiles (n=4) on "Total Monthly Spend" give Lowest (Q1), Lower-Mid (Q2), Upper-Mid (Q3) and Highest Spenders (Q4), which are then crosstabbed against brand preference.
-   **Benchmarking and ranking:** set performance benchmarks by splitting the population into top and bottom performers. Example: Deciles (n=10) on a CSAT score to find the Top 10% (10th decile) and Bottom 10% (1st decile) and compare their characteristics, feedback or demographics.
-   **Handling skewed data:** each segment has an equal number of observations, so comparison groups stay reliable even when the raw distribution is irregular (income, website visits).
-   **Simplifying analysis and reporting:** reduce a continuous scale to a few categories (for example three Tertiles instead of ages 18 to 65).

**Types**

| N-Tile type | Definition |
| --- | --- |
| Tertiles (n=3) | Three equal parts |
| Quartiles (n=4) | Four equal parts (Q1 25th percentile, Median 50th percentile, Q3 75th percentile) |
| Quintiles (n=5) | Five equal parts (20% intervals) |
| Deciles (n=10) | Ten equal parts (10% intervals) |
| Percentiles (n=100) | 100 equal parts (1% intervals) |

**How to create**

-   **Crosstab Settings:** Crosstab Settings > N-Tiles > select targets and quantile type > **Generate**.
-   **Coding Grid:** Coding Grid > kebab dropdown > N-Tiles > configure and generate.

## **Volumetrics**

The Volumetric modal applies mathematical aggregations to numeric variables **before** table insertion.

-   Calculations: **TOTAL, MEAN, MEANZ, MEDIAN, MEDIANZ**.
-   Output: combined into one statement, or separated across individual rows / columns.
-   Use **DEC()** syntax in statement options to set decimal precision.

**Process**

1.  **Select codes:** tick the codes, categories or subcategories in the Codebook.
2.  **Initiate addition:** click the **Add To** button for rows, columns or tablebases.
3.  **Choose method:** pick **Volumetric**; this opens the Volumetric modal.
4.  **Define calculations and output:** choose the calculations (TOTAL, MEAN, MEDIAN and so on) and whether results are combined into one row / column / tablebase, or calculated separately with each code on its own line.

The modal also has statement titles and coding previews, so you can name the resulting statement and see the code that will be applied. Click **Generate** to create the row, column or tablebase item.

## **Add to Multiple**

Maps selected Codebook items en masse across multiple target items at once. It is an extra method for moving items from the Codebook to the crosstab and simplifies applying selected survey answers or codes to several pre-existing items.

1.  **Selection:** choose codes, categories or subcategories in the Codebook.
2.  **Initiation:** click the **Add To** buttons (rows, columns or tablebases).
3.  **Method choice:** select **Add to Multiple** (the other methods are Separate, Combined, Volumetric and Count). This opens the "Add code to multiple" modal.

## **Large Row Analysis (LRA)**

Explore supports a maximum of **5,000 rows and 150 columns**. LRA handles high-density categories (selecting categories / subcategories with up to **50,000 items**) by filtering down to focal subsets.

1.  Launch LRA from table settings or the Coding Grid kebab menu.
2.  Select target categories and the number of **Top** or **Bottom** rows to isolate.
3.  Choose the sorting metric (for example %Col) and apply index thresholds.
4.  Preview the results and choose to **append** to or **overwrite** the current crosstab rows.
5.  Explore passively saves LRA configurations across consecutive runs in the active session.

**Not supported:** volumetric items, means, medians and totals are not supported in LRA. This is by design. Users only need to know these items are unsupported; no workaround is needed.