Topics In Demand
Notification
New

No notification found.

SEEING IS BELIEVING: KEY GUIDELINES FOR DATA ANALYSIS AND VISUALIZATION (1/2)
SEEING IS BELIEVING: KEY GUIDELINES FOR DATA ANALYSIS AND VISUALIZATION (1/2)

April 13, 2022

284

4

This is the third and final blog in the series of ‘Best Practices in Crafting Effective Surveys, Data Analysis and Visualization’. For ease of reading, this blog has been split up into two parts. While the first (current one) covers Data Analysis and Visualization for CATEGORICAL/NOMINAL VARIABLES, the second part will encapsulate the analysis and presentation methods for SEQUENTIAL/ORDINAL and INTERVAL/RATIO types of variables.

Presentation is key. In fact, the breakneck speed at which today’s world is transforming, one can even say presentation and packaging is (nearly) everything. In our lives, most of us would have come across popular adages like, ‘first impression is the last impression’ and ‘what you see is what you get', and a lot of times we would have experienced the truth in these aphorisms as well. One can prepare the most delectable dish, but if it lacks visual appeal, in all probability, no one would be willing to even taste it. The same goes for data. One can perform the most complex calculations through intricate models while analyzing data, but at the end of the day, if what goes out on the table is not palatable and coherent, all the good work at the analysis stage will fall by the wayside. The presentation/visualization of data reflects the manner in which one wishes to communicate with their audience/readers, and thus, it is imperative that data is presented in a clear, succinct, cohesive, and comprehensible manner. My second blog elucidated how the wide variety of question types can not only add value to our survey design but also increase the response rate for survey questionnaires. In my final and concluding blog, we will delve into some key best practices that aid the analysis, interpretation, and presentation of complex data, collated through survey questions, in an ‘easy-to-digest’ format.

The type of questions and variables being considered for the analysis of responses will feed into the course of analysis. Hence, before we step into the data analysis/ presentation bit, my suggestion to the readers is to kindly go through my previous blog: https://community.nasscom.in/communities/impact-stories/how-ask-potpourri-various-survey-question-types for a detailed understanding of major question types, based on the underlying variables under consideration, which can be used while drafting survey questionnaires. Enumerated below are some recommended methods of analyzing and presenting data in a lucid and comprehensive manner, based on the nature of the questions and variables being analyzed:

  1. CATEGORICAL/NOMINAL VARIABLES:

As explained in my previous blog, these variables can be categorized into two or more groups, with no intrinsic ordering of the categories. Analysis of the data which studies these variables can be done using simple counts and percentages (adding up to a total of 100), as well as by computing averages, median, percentiles, etc., and testing correlations as well. Nominal variables can be aptly studied through closed-ended questions, such as -

  1. Dichotomous (‘Yes/No’), Radio Button (Single Answer), Dropdown (Single Answer), and Matrix/Grid (Radio Button/Dropdown – Single Answer)

SUGGESTED OUTPUT/REPRESENTATION

  • Calculate percentages, which will total to a 100% and present in the form of a PIE/DOUGHNUT (DONUT) CHART OR a STACKED BAR (HORIZONTAL) CHART
ABCD

 

ABCD

 

NOTE: In case of economic indicators, such as GDP growth, inflation, etc., asked as single-answer timeline questions (covering sequential time periods), such as -

Q   What was your company’s growth rate in 2018, 2019, 2020, and 2021?

  • If this is an open-ended question type, then one can average out the responses of growth rates per year, for 2018, 2019, 2020, and 2021 and then show the data in the form of a LINE CHART. Illustratively,
ABCDbcgh
  • If the above question is asked as a single-answer question type but with predefined answer options across multiple years, then one can total the responses for each range, calculate the percentage of each answer option and since the total will be 100%, one can present it as a STACKED BAR/COLUMN CHART (Stacked bar chart can be created in the same manner as shown above in this blog). For instance –

Q   What was your company’s revenue growth rate in 2018, 2019, 2020, and 2021?

  •  <5%
  • 5-10%
  • >10%

This question can be analyzed as shown in the tables below:

bcghbcgh

Also, please note for representing averages, medians, percentiles, etc. one can use a STOCK CHART as illustrated below:

ABCD

 

b. Multiple Choice (Many Answers)/Checkbox/Dropdown (Multiple Answers) and Matrix/Grid (Checkbox/Dropdown – Multiple Answers)

SUGGESTED OUTPUT/REPRESENTATION

  • For analyzing these questions, count each answer option selected, as one response, divide it by the total number of responses received (all the people who have answered the question), and accordingly calculate percentages. For example, in the table below Respondent 1 has ticked 3 answer options (Vanilla, Butterscotch, and Strawberry), so that would count as 3 responses in the numerator but 1 response in the denominator. The sum of the answers will not total to 100%. The answers can then be presented as a BAR/COLUMN CHART.

Q Which of the following ice-cream flavours do you like:

bcgh

ANALYSIS:

Percentage of respondents preferring each ice-cream flavour:

  • Vanilla - 3/5=60%
  • Chocolate - 4/5=80%
  • Butterscotch - 2/5=40%
  • Strawberry - 2/5=40%
  • Mango - 1/5=20%
  • Others (Coffee) - 1/5=20%
ABCDABCD

Even though in today’s day and age we have numerous highly intuitive tech platforms, be it PowerPoint, Prezi, Canva, Visme, Tableau, Power BI, etc., or the wide array of infographics available on the Web, which offer novel ways of analyzing, presenting and visualizing data, it is still imperative to ‘get the basics right’ to be able to deploy and utilize the right data analysis and visualization tool.

To understand the preferred analysis and presentation/visualization methods for SEQUENTIAL/ORDINAL and INTERVAL/RATIO types of variables, please read the second part of this blog coming up in the next few days.


That the contents of third-party articles/blogs published here on the website, and the interpretation of all information in the article/blogs such as data, maps, numbers, opinions etc. displayed in the article/blogs and views or the opinions expressed within the content are solely of the author's; and do not reflect the opinions and beliefs of NASSCOM or its affiliates in any manner. NASSCOM does not take any liability w.r.t. content in any manner and will not be liable in any manner whatsoever for any kind of liability arising out of any act, error or omission. The contents of third-party article/blogs published, are provided solely as convenience; and the presence of these articles/blogs should not, under any circumstances, be considered as an endorsement of the contents by NASSCOM in any manner; and if you chose to access these articles/blogs , you do so at your own risk.


images
Prerna Buckshee
Manager - Research

© Copyright nasscom. All Rights Reserved.