gym-original.png
gym-redesign.png

Parking Sign Comprehension - Text + Graphics

Experimental Research Methods - Master's Program

Class Objectives

•    Design and execute a research experiment
•    Construct a research problem and specific research question based on relevant literature
•    Identify relevant variables
•    Create hypotheses
•    Design a research study
•    Construct and administer experimental instruments
•    Select and apply appropriate data analysis methods
•    Write a research report
•    Evaluate the validity of experimental research studies
•    Identify and understand ethical concerns with experimental research online

Individual Responsibilities

  • Helped develop study methodology
  • Created recruitment tools
  • Designed and built survey in SurveyGizmo
  • Facilitated data collection
  • Contributed to final report content and design

 

Background

Lack of clarity in parking signs is causing frustration and anger among drivers across cities in the US. Often, misinterpretation of these signs has led to avoidable parking citations for drivers and increased administrative costs for cities (TicketZen.com). In 2014, the Department of Transportation (DOT) in Los Angeles generated an estimated $106 million in revenue from parking violations alone, and authorities admitted that a substantial portion of this revenue was derived from confusing parking signage.

The problem with comprehending standard US parking signs is that they are dense with text, full of information that is necessary for driver compliance of parking laws. This lies contrary to previous research and the basic tenets of Visual Information, that text-only presentations are not as effective in reader comprehension as text accompanied by images (Gershon et al. 1998). Therefore, a re-imagining of the standard design has been proposed to help alleviate the problem drivers currently have with street parking.

 Nikki's work on redesigning the current parking signs.

Nikki's work on redesigning the current parking signs.

Brooklyn-based designer Nikki Sylianteng (http://nikkisylianteng.com/) introduced a design that visually presented the same information contained in existing parking signs but in timeline form. The city of Los Angeles, in their quest to alleviate driver frustration, recently implemented a pilot program that would replace the familiar strip of interrelated parking signs with Sylianteng’s single, consolidated, grid-like design (LADOT Parking Sign Pilot Program 2015). 

We wanted to determine whether comprehension is tied to the type of presentation the information is conveyed in. Therefore, we investigated the effects of the visual-textual information presentation and the existing text-only information presentation on the efficiency of comprehension of US street parking signs.

Problem Statement: This study investigates the effect of visual and textual information on the comprehension of US street parking signs.

Hypothesis: Parking signs that include both visual and textual information increase the accuracy and efficiency of comprehension in drivers than the current design of using text alone.

 

Study Methodology

Screener Questions

We conducted an online between studies design survey staged as a comprehension test of highway signs. First we asked screening questions to make sure participants were

  • Over 18 years old
  • Have a US driver's license (consistency)
  • Have driven at least once in the last month (familiarity)

Randomized Sign Questions

The study required participants to review a series of common traffic signs (distractors) for comprehension, along with the parking sign, allowing the study to be conducted as a general driving test. Participants were asked to review a total of four signs, and answer three questions after each sign was displayed – one question to measure sign comprehension and two questions to measure satisfaction with the sign.

For the parking sign question, the study compared a current street parking sign with a graphical redesign of the same sign. Participants in the control group were shown the current street parking sign and participants in the treatment group were shown the redesign, both shown below.

control_treatment_signs.PNG

Post test questions

At the end of the main study questions, participants were asked a few questions for us to better understand our results. These questions were asked last to avoid leading the participants to focus on the parking sign question.

  • Street parking in the last month (moderating variable)
  • [If in the treatment group] Had they seen the redesigned sign before (disqualifier)

At the very end of the survey, the participants were completely debriefed on the study goals and were able to contact our team if they wanted their responses to be deleted from the rest.

Survey Data

We collected data on the following:

  1. Accuracy of response (categorical variable)
  2. Response time (continuous variable)
  3. System Usability Scale (SUS) questions on ease of use and confidence in the signs information (Likert Scale, Avg.- Interval variables)
 Questions asked for each sign, including the treatment and control parking signs.

Questions asked for each sign, including the treatment and control parking signs.

Data Analysis

Statistical Tests

In order to measure the differences between the treatment and control groups, we plan on performing four statistical tests.

  • An independent T-test to determine whether there is a statistically significant difference between the means of the response time in the two groups.
  • A 2x2 Chi-Square test to measure and compare the accuracy of response between the two groups.
  • For user satisfaction and confidence, we will take the average of the two SUS responses and perform a T-test on that average.
  • Experience with street parking, age, gender and type of platform used to respond to the survey will be tested for their effect, if any, on the response times with a 2-way ANOVA.

Results

If you want to see the results of our study, please contact me for the full report using duffymp9@uw.edu.

The full report includes the following additional information:

  • Full results of each statistical test
  • Study limitations
  • Design implications
  • Future work and recommendations