UX Analytics

A Social Network Analysis academic project​ on creating data-driven design

Project’s Problem Statement

The problem states that, currently, on the Google Maps application, the cancel button is positioned in the bottom-left corner. UX researchers have found that users would feel more accessible to find the cancel button on the top-right corner based on user interviews. The researchers want to confirm this design based on empirical data to determine the UI reform.

How would you design the experiment and analyze data?

Data set was provided. 

 

Team:

Solo

Tools:

IBM SPSS

 
Research Question
What location (Bottom Left or Top Right) of the cancel button is much easier for the Google Maps users to find? 

Creating a Hypothesis

The experimental design begins with formulating a hypothesis —an educated guess about causation.

  • A hypothesis must not only be testable, it must also be falsifiable.
  • We can’t actually prove a hypothesis to be true: we can only show that it is false.
  • What is really being tested is an ‘anti-hypothesis’: an assertion that the independent variable will have no effect on the dependent variable.
  • If we can disprove the anti-hypothesis then we may conclude that our original hypothesis is correct. (In experimental design, the ‘anti-hypothesis’ is known as the ‘null hypothesis)

My Hypothesis

My hypothesis for the problem is that the Google Maps users will take much less time to find the cancel button if it is on the Top Right side.

Defining the variables for our experiment:

Independent Variable

The location of the cancel button (Xlocation) on the Google Maps application, i.e., bottom-left / top-right

Dependent Variable

The time taken (TimeTaken) by users to find and click on the cancel button

Experiment Design

For this experiment, I will chose ‘Between-subject design’. In order to avoid order effects and obviousness of the experiment, I will assign different participants to view different locations of the cancel button.

What is ‘Between-subject design’?
Between-subjects is a type of experimental design in which the subjects of an experiment are assigned to different conditions, with each subject experiencing only one of the experimental conditions. 

Applying Statistical Concepts​

  • If we have an unrelated predictor variable comparing two groups and an outcome measured on a continuous scale, we need to analyze our data with an independent samples t-test.
  • If we have a continuous predictor variable and a continuous outcome measure, we need to perform correlation analyses.
  • If we have a categorical predictor variable and a categorical outcome measure, we need to perform a chi-square.

SPSS Tables for Descriptive Analysis

SPSS Tables for Inferential Analysis

Interpretation and Summary

  • It is indicated by the analysis that when the cancel button is located on the bottom-left, the time taken for users to find and click on the cancel button ( 8 min 50 sec ± 2 min 17 sec) is statistically significantly slower compared to when the cancel button is located on the top-right (7 min 7 sec ± 1 min 40 sec ), t(433) = 9.05, p < .001
  • The hypothesis is supported that the users will take much less time and find it easier to locate the cancel button if it is located on the Top-right position in the Google Maps application.

Interested in collaborating with me? Or, have any feedback? I’d love to connect with you!

Email: Shubhangislal@gmail.com

Download Resume