This project focused on designing an experiment to evaluate the impact of using physical keyboards versus touchscreen devices on typing efficiency. The goal was to develop a structured methodology that would allow for the collection of meaningful data, providing insights into how different text input methods influence user performance and experience.

The experiment was designed to test the hypothesis that physical keyboards enable faster and more accurate typing compared to touchscreen devices due to their tactile feedback. To ensure a fair comparison, a controlled experimental setup was established, accounting for variables such as participant demographics, prior typing experience, device specifications, and environmental conditions.

The study incorporated both quantitative and qualitative data collection methods. Typing performance was measured in terms of words per minute (WPM) and accuracy using a standardized typing test, while a participant satisfaction survey captured subjective experiences related to comfort, ease of use, and perceived efficiency. A counterbalancing approach was used to mitigate order effects, ensuring that participants completed typing tasks on both input methods in a randomized sequence.

May 2024
Cardiff University- Group Project 

Design an experiment to assess text input speed

Aim-

To evaluate the impact of using physical keyboards versus touchscreen devices on typing efficiency, providing insights for user-cantered design in text input technology.

Hypotheses-

H1 (Alternative Hypothesis): Participants will type a given passage with a higher speed (words per minute) on a keyboard compared to using a touchscreen device, due to the tactile feedback provided by physical keyboards, which can enhance typing accuracy and speed in contrast to the less tactile, flat surface of touchscreens.

H0 (Null Hypothesis): The ratio of typing speeds will be equivalent, indicating no significant difference in typing speed between using a keyboard or using a touchscreen device.

Variables-

Independent Variable (IV): The type of text input method.

Levels:

  1. Keyboard (Physical Input)

  2. Touchscreen (Touchscreen Device)

Dependent Variable (DV): Net Typing Speed (Words per Minute)

Control Variable:

  • Participants-

    • Native Language: English fluency and native language can impact typing speed due to familiarity with the language and keyboard layout.

    • Age: Younger individuals might have faster typing speed and better hand-eye coordination, but older individuals might have more typing experience.

    • Experience: Individuals who type regularly for work or leisure (e.g., writers, programmers) are likely to have faster typing speeds than those in non-typing-intensive professions or fields of study.

  • Equipment:

    • Keyboard Layout and Design: Different layouts (QWERTY, AZERTY, Dvorak) and designs (mechanical vs. membrane, key size, space between keys) can affect typing speed.

    • Touchscreen Type: Capacitive vs. resistive touchscreens can impact responsiveness and accuracy, affecting typing speed.

    • Screen Size and Resolution: Larger screens may improve typing accuracy on touchscreens, but too large can also hinder speed due to increased finger travel distance.

    • Keyboard and Screen Condition: Worn-out keys or a scratched and unresponsive screen can slow down typing speed and reduce accuracy.

  • Physical Environment and External Factors:

    • Desk and Chair Ergonomics: Proper typing posture facilitated by comfortable and ergonomically designed furniture can impact typing speed and accuracy.

    • Lighting: Poor lighting conditions can strain eyes, leading to slower typing speed and increased errors.

    • Noise Level: High noise levels can distract typists, potentially slowing down their typing speed and reducing concentration.

  • Psychological and Physiological Factors:

    • Mood and Stress Levels: High stress levels or a bad mood can negatively impact typing speed and accuracy.

    • Fatigue: Tiredness can significantly reduce typing performance and attention to detail.

    • Health Conditions: Conditions such as carpal tunnel syndrome, arthritis, or any hand injuries can severely affect typing ability.

Data Collection-

1. Typing Performance:

Metric: Use TypingTest.com (online webpage) to calculate Net Speed (Words Per Minute, WPM) calculated as Typing Speed (WPM) multiplied by Accuracy (Percentage of Correct Words, %).

Type of Data: Discrete Quantitative data. This data will later be converted into a ratio for analysis.

2. Participant Satisfaction Survey:

Metric: Responses to Likert-scale questions and qualitative feedback.

Type of Data: For Likert-scale questions: Ordinal. The responses can be ranked (1 being ‘Strong Disagree’ and 5 being ‘Strong Agree’). For open-ended questions: Qualitative. These responses will provide narrative data that can give context to the quantitative findings and reveal deeper insights into user experiences.

Experiment Procedure-

Participants will be evaluated on their typing speed and accuracy using TypingTest.com. This platform is selected for its realistic simulation of real-world typing scenarios, wherein users are tasked with typing coherent passages, that they can see more words as they type. This approach closely mirrors actual typing activities, as opposed to alternative methods that measure performance based on the typing of randomly generated words. TypingTest.com offers the added advantage of automatically calculating both the typing speed (WPM) and the accuracy percentage.

  1. Environment and Setup: Conduct all tests in a quiet, well-lit room. Plan all experiments between 10 am and 12 pm to reduce the influence of circadian rhythm. Establish two workstations, one equipped with the keyboard and the other equipped with touch screen. Prepare both the devices with Typingtest.com with the 3-minute medium test. Use ergonomic chairs, give participants five minutes to adapt to the setup, make sure they can touch all the buttons comfortably. Take a 5-minute break between tests to reduce fatigue.

  2. Ethics and Confidentiality: Obtain the informed consent of all participants, detailing the research objectives, procedures, data processing, anonymity, and rights, including withdrawal excluding consequences. Standardized presentations can ensure that all participants get the same information without technical bias.

  3. Matching Participants and Grouping: Participants will be stratified based on occupation, age, and native language. This stratified sampling approach ensures that a wide range of users is represented, potentially uncovering nuanced insights into how different demographic and occupational groups interact with these technologies. Occupation: ‘typing-intensive’ occupations (e.g., writers, secretaries, programmers) and ‘non-typing-intensive’ occupations (e.g. contractor, builder, sales assistant). Age: 18–30, 31–45, 46–60, 70+. Native Language: Native English Speakers and Non- Native Speakers. To gather a diverse sample of participants, employing a multifaceted approach, utilizing online forums, social media, and community outreach will ensure wide demographic coverage. Participants will be screened to classify them into ‘typing-intensive’ versus ‘non-typing-intensive’ occupations and further categorized by age and native language proficiency. This will create 12 subgroups. Then randomly selection participants within each subgroup to minimize bias, choosing the number of participants for each subgroup based on what is reflects the wider population. Altogether there’s should be 1000 Participants as sample sizes should be 10% of the wider target population without exceeding 1000. Participants are then randomly divided into Group A and Group B, as they will do opposite tasks first for counterbalancing. Participants for each subgroup must be evenly divided to Group A and Group B using a random generator.

  4. Each participant will undergo a short practice session on the keyboard and touchscreen device before starting the test. This is to minimize any unfamiliarity with the testing platform or the typing interface. Each participant will complete two typing sessions, one using the keyboard and one using the touchscreen. Group A participants will use the keyboard first and Group B participants will use the touchscreen first. All participants will use texts of the same difficulty level (Medium) to ensure comparability. Different texts will be used for the iPad and the laptop sessions to reduce learning bias. This ensures that improvements in typing speed or accuracy are not due to memorization of the text. Each typing session will be timed to last exactly three minutes, standardizing the duration. Immediately following each typing test, the software will provide metrics for typing speed (WPM) and accuracy (percentage). These metrics will be recorded for further analysis.

  • Participant Satisfaction Survey: The satisfaction survey will be administered immediately after participants complete both typing tests to ensure that their experiences are fresh in their minds. This survey aims to capture their subjective experiences and preferences when typing on both the physical keyboard and touchscreen device. Participants will be asked to rate questions on a scale of 1 to 5, with 1 being ‘Strong Disagree’ and 5 being ‘Strong Agree’. Definitions for each rating level will be clearly provided to ensure consistency in participant responses. The survey will include questions targeting specific aspects of the typing experience, such as comfort, ease of use, perceived speed, and error frequency. This approach helps identify strengths and weaknesses of each input method. In addition to structured rating questions, the survey will feature open-ended questions allowing participants to elaborate on their experiences, express preferences, and provide suggestions for improvement. Participants will be informed that their responses are anonymous to encourage honesty and reduce bias. The survey will be conducted in the same environment as the typing tests to maintain consistency in participant experience and mood. Ratings will be averaged for quantitative analysis, while qualitative responses will be categorized and analysed for common themes and insights.

  • Data Handling and Privacy: The data is anonymized using a unique participant code. Using strict data protection and follow the university guidelines on data retention and processing after research.

  • Concluding Procedures: After the meeting, thank each participant and make sure that all your data have been backed up and saved correctly. Prepare the workstation for the next participant and reset all parameters and maintenance equipment including sanitary conditions and standards.

  • Data Analysis: To start analysis of data, use descriptive statistics (mean, median, standard deviation, range) of the dependent variables (typing speed and accuracy) for each group (keyboard and touchscreen).