Introduction to Descriptive Analysis

Before you look at consensus, sentiment, or subgroup differences, you need to understand the basic patterns in your data. That's where descriptive analysis comes in.

Descriptive analysis means:

• Summarizing
• Organizing
• Displaying

…your Delphi results so you can see clearly what experts think.

Why Use Descriptive Statistics?

Descriptive statistics are the foundation of Delphi reporting because they:

• Show where most participants place an item (central tendency)
• Reveal how much opinions vary (dispersion)
• Help you rank items by priority

In other words: They make complex expert opinions digestible and comparable.

Core Elements of Descriptive Analysis

Let's break it down step by step:

1. Central Tendency

What is it?

Measures that show the "typical" response.

Mean (Average):

Sum of all ratings divided by the number of ratings. Can be influenced by extreme values.

Example:

Scenario: Universal Basic Income by 2030.

Ratings (1–5): 3, 3, 4, 4, 4, 5, 5

• Median = 4
• Mean = 4

Interpretation: Strong central tendency around 4.

2. Dispersion (Variability)

What is it?

How much do responses differ?

Key metrics:

IQR (Interquartile Range):

• Q3 minus Q1
• Shows the spread of the middle 50% of ratings
• Good for consensus assessment

Standard Deviation (SD):

• Average distance of each rating from the mean
• Reflects overall variability

Example:

Ratings (1–5): 2, 2, 3, 4, 5

• Median = 3
• IQR = 2
• SD ≈ 1.14

Interpretation: Opinions are quite dispersed.

3. Item Ranking

What is it?

A way to prioritize items e.g., by their mean score.

Why do it?

• Helps you identify the most supported scenarios or priorities.
• Makes it easy to show what experts think matters most.

Example:

Items ranked by median:

• Scenario A: Mean = 5
• Scenario B: Mean = 4
• Scenario C: Mean = 3

Interpretation: Scenario A is top priority.

Example Use Case

Imagine your Delphi focuses on future public health innovations. After descriptive analysis, you find:

Interpretation:

• Predictive analytics: highest agreement and top priority
• Blockchain: more divided opinions
• AI diagnostics: strong but slightly less consensus

Why Is Descriptive Analysis Important?

• First step of interpretation — all other analyses build on this.
• Quickly shows:
  • What's popular
  • What's controversial
  • Where consensus might exist
• Provides transparent evidence for reports and publications.

How Durvey Helps

Durvey automates your descriptive analysis:

• Calculates mean, IQR, SD for each item.
• Sorts items by any metric (e.g., mean descending).
• Visualizes distributions clearly.
• Exports paper ready results for further processing or publication.

No expertise required.

Example Interpretation for a Report

"Descriptive analysis showed that predictive analytics was rated highest (mean 5, IQR 0.5), indicating strong expert support. In contrast, blockchain health records received a lower mean (3) and higher dispersion (IQR 2), reflecting less consensus."

Durvey lets you visualize and communicate these insights instantly — so you can focus on what matters most: interpreting the data and making informed decisions.