User Centered Design and
Proceedings
Editor Editorial board
McGraw Hill Education (India) Private Limited NEW DELHI
New Delhi New York St Louis San Francisco Auckland Bogotá Caracas Kuala Lumpur Lisbon London Madrid Mexico City Milan Montreal San Juan Santiago Singapore Sydney Tokyo Toronto
iii
Published by McGraw Hill Education (India) Private Limited, P-24, Green Park Extension, New Delhi 110 016. Proceedings International Ergonomics Conference Copyright © 2014, by McGraw Hill Education (India) Private Limited. No part of this publication may be reproduced or distributed in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise or stored in a database or retrieval system without the prior written permission of the publishers. The program listing (if any) may be entered, stored and executed in a computer system, but they may not be reproduced for publication. This edition can be exported from India only by the publishers, McGraw Hill Education (India) Private Limited. ISBN (13): 978-93-392-1970-3 ISBN (10): 93-392-1970-8 Managing Director: Kaushik Bellani Publishing Manager – Professional: Mitadru Basu Development Editor – Science Technology & Computing: Deepika Jain Manager—Production: Sohan Gaur General Manager – Sales and Business Development: S Girish General Manager—Production: Rajender P Ghansela Manager—Production: Reji Kumar
Information contained in this work has been obtained by McGraw Hill Education (India), from sources believed to be reliable. However, neither McGraw Hill Education (India) nor its authors guarantee the accuracy or completeness of any information published herein, and neither McGraw Hill Education (India) nor its authors shall be responsible for any errors, omissions, or damages arising out of use of this information. This work is published with the understanding that McGraw Hill Education (India) and its authors are supplying information but are not attempting to render engineering or other professional services. If such services are required, the assistance of an appropriate professional should be sought. Typeset at Script Makers, 19, A1-B, DDA Market, Paschim Vihar, New Delhi, and printed at Cover Printer: Cover Designer: Kapil Gupta
iv
A study on efficiency of input devices on native language numerical data entry Shrikant Salve*, Pradeep Yammiyavar** Department of Design, Indian Institute of Technology, Guwahati, India, *
[email protected], **
[email protected] Abstract This study focuses on the efficiency of input devices like keyboard and mouse on numerical data entry in the contexts of semi urban or rural users from India. We address the question whether the input devices like keyboard and mouse differentially influence performance in native language numerical data entry. Numerical data entry is being done at various places in rural and semi-urban area of India. The user uses either keyboard or mouse for numerical data entry. A study was conducted using an experimental interface designed to examine the efficiency in terms of time required and errors made in native language (Marathi or Assamese) numerical data entry. Subjects from semi urban and rural India were given a task to perform. The task involved use of a calculator which was specifically designed for this experiment. This calculator takes input from keyboard as well as mouse in native languages. To calculate the efficiency, the computer based background recording of the data input has been taken. The results show that there is significant difference in use of input devices (keyboard and mouse) for native language numerical data entry by rural users. Keywords – numerical entry errors, native language, input devices keyboard and mouse while numerical data entry done by rural or semi urban users.
1. INTRODUCTION Human being often make errors such as mistakes, slips, lapses, miscalculations etc. in their interaction with computers. Even though accepting error is common, their effects on the human being can range from being frustrated to the disasters depending upon the context of the data. A task of feeding information in a computer has many types of errors called as data entry errors. Simple data entry errors such as typing an incorrect number, typing a number twice or skipping a line can give wrong results. In rural areas of India computers are used by people in many places like banks, railways, bus stands, hospitals, factories, government offices, market places / shops and NGOs (data entry jobs). In such a context data entry errors can have harsh effects on rural people who use computers in to earn a living. Many interactive computing devices involve entry of numerical data that represent drug doses entry in medical field, ATM numerical data entry etc. In rural India computers are used by people in many places like banks, hospitals, shops etc. for numerical data entry [5]. Number entry is a task that is performed daily with less conscious thought. These number entry tasks seem unimportant, but we still experience problem when performing them like typing an incorrect number or typing a number twice and many more. It has been observed that people enter wrong number in number entry fields like account number, amount, PIN etc. It rare to enter wrong number, but it does happen occasionally and the results are potentially embarrassing or even disastrous. All these system use a simple input method, such as keyboard and mouse. Therefore, our study aimed towards finding the efficiency of input devices like
167
2. RELATED LITERATURE This section details literature that has influenced this study of efficiency of input devices on number entry task. What is Error? James Reason [4] defines two most important categories of errors such as slips / lapses and mistakes. Slips results from the incorrect execution or incorrect planning of a correct sequence of actions, so error is made even though a person has the knowledge needed to perform the task. In contrast, mistakes occur when a person has incorrect or no knowledge of the task they are aiming to complete. In other words, mistakes result from the correct execution of an incorrect sequence of actions. For example, consider the misinterpretation of number: Reading “161” as the expected value “16.1” could be slip, whereas thinking a symbol signifies “plus” when it signifies “plus-minus” is probably a mistake. There are several places [5] in India where the numerical data entry has been done which also include NGOs. Singh et. al. [6] have proposed study on the numeric paper forms used by the NGOs for data collection in rural India. They have investigated NGO’s form filling requirements which were used to interact with rural people. They proposed the numeric input method for different NGO’s form filling requirements which is easy to use for rural people and also machine readable. This paper's context is the data entry jobs provided for rural people by various NGOs in India.
168 Literature states few studies conducted on numerical data entry. Barchard et. al. [1] have projected the study on impact of human numerical data entry errors on statistical results and calculations. They used 195 undergraduate students to participate in experiment by assigning three data entry methodsdouble entry, visual checking and single entry to them. The participants entered 30 data sheets, each containing six type of data. Their results show that in double entry significantly fewer errors than both visual checking and single entry. Another study proposed by Oladimeji et. al. [3] of number entry interface found on medical devices. They reported an experiment that investigates the effect of interface design on error detection in number entry tasks using two number entry interfaces, one serial interface with 12 key numeric keypad and another incremental interface that use a knob or a pair of keys to increase or decrease numbers. 22 participants aged 18-55 years took part in the experiment. A computer was used with an integrated eye-tracker to present the instructions and number entry interfaces. Each participant used both number entry interfaces (independent variable). The dependent variables were the number of undetected errors, number of corrected errors, total eye fixation time and task completion time. The participant was required to enter 100 numbers using both interfaces according to the instruction shown on the right half of the screen. The fixation experiments showed that users are less likely to notice errors on the display when using keypads, as they fixate twice as much on the keypad itself compared to the up/down buttons. They also identified six categories of number entry errors (skipped, transposition, wrong digit, missing decimal, missing digit and other). They have suggested giving priority to research number entry styles and their relation to error rate, behavior and performance in the context of safety critical number entry systems. However their study restricted to medical number entry systems. One more study on number entry reported by Eslambolchilar et. al. [2] for telephones. They analyzed errors associated with three types of digit entry such as rotary dial, push button keypad and touch based input. Thirty participants took part in experiment of eleven digit number entry by using above three digit entry interface of telephone. The results concluded that the touch interface was four times less accurate than the pushbutton and rotary dial interface. This study limited to telephone digit entry which does not contain decimal point. We focus our research on input devices in numerical data entry and the efficiency in terms of error and time associated with it.
3. EXPERIMENTAL SETUP Current literature studies on number entry research mainly focus on a specific medical domain. The focus of this study was more on everyday interaction with
computer for numerical data entry at banks, government offices, BPOs by rural and semi urban Indian users. This is the type of numerical entry they perform almost on daily basis, for example entering account number or amount or PIN, where one wrong digit invalidates the whole results which are potentially embarrassing or even disastrous. Therefore, consideration of the numerical data entry errors are crucial for evaluating efficiency of rural users while interacting with computers. We are investigating the influence of input devices in making these errors. The user study was performed to compare the both input devices such as keyboard and mouse. A total of forty eight participants (M = 20.89 years; SD = 3.07; range = 18-30 years) (male / female) working in the IIT Guwahati (IITG) campus that is – people working at shopping complex, vegetable market and security guards as well as and 24 rural villages based users from Maharashtra India were selected. Participants had educational qualification of 10th to 12th standard (that is non-graduate) and used computers or laptops at least one hour in a week. The following figure 1 depicts the participants performing given experiment.
Fig 1. User’s participation in experiment, above two pictures showing numerical data entry by using mouse and below two by keyboard as input device. A software interface was designed specifically for this experiment. It was designed to take numerical entry input in English and two native languages (Marathi and Assamese) using input device such as mouse and keyboard. But only native languages interface were used for this experiment. This software interface does arithmetic operations such as addition, subtraction, multiplication and division on numerical with and without decimal point. Figure 2 depicts the three screen shots of English, Marathi and Assamese language calculator respectively.
169 errors. Table 1 summarizes the design of tasks form T1 to T4 as, T1- input device mouse and without time limit and likewise. Table 2 depicts the total number of digits that were incorrect in each task. Table 1. Tasks design Fig 2. Screen shots of software interface designed, which does calculations in three languages – English, Marathi and Assamese The experiment was a ‘within subject repeated measures’ design. The participant used software interface to perform tasks for calculation in both native languages. The participants perform same calculation in both native languages using two input devices mouse and keyboard. The input devices was the independent variable and it had two types: keyboard and mouse. The dependent variables were the task completion time and errors made in numerical data entry. All participants were tested individually. The software interface was used for each participant and they were briefed about the stages and purpose of the experiment before starting. The experiment was divided into four parts (tasks), for numerical entry by keyboard and mouse within and without time limit. We have taken mixed native language participants, twenty Assamese language speakers and twenty eight Marathi language speakers. Each participant has to perform all tasks, but the sequence/order of the tasks may be different. Prior to each stage of the experiment, the participants were given orientation session where they could enter 2/3 simple calculations and get familiar with the interface. When the participants were comfortable with how the interface worked, they were allowed to proceed to the experiment. The participants were required to enter given mathematical calculations having three different difficulty levels (like very easy, easy and hard) using given interfaces (Marathi or Assamese) in the defined order. The participants were provided the experiment sheet including mathematical calculations in native language they speak. The participants were instructed to perform the mathematical calculation as quickly and as accurately as possible. The computer based background recording of each participant interaction with designed software interface have taken to collection of speed of entry and errors.
Input Without Time Device Limit Mouse T1 Keyboard T2
Within Limited Time T3 T4
Table 2. Total incorrect digits and error rate for respective tasks Task Total Incorrect Digit T1 14 (1.56) T2 21 (3.25) T3 24 (2.72) T4 26 (4.11)
Table 3 shows the total number of each type of error on each input device, as well as the frequency of errors on the input devices. Here ‘n’ represents the number of errors made and ‘r’ is the frequency. The frequency is calculated as ‘100n/N’ percent for that particular error occurred on that input device. A total of 38 digits were entered incorrectly by mouse, considerably less than the keyboard. The most common errors by both input devices were entering incorrect / wrong digit and missing digit. Table 3. Mouse Errors (M), Keyboard Errors (K) Error Type
nM
Decimal point Missing digit Wrong digit Double entry Reverse digit
6 8 20 3 1
rM (N=1817) 0.34 0.45 1.13 0.17 0.06
nK 8 15 16 5 3
rK (N=1839) 0.63 1.17 1.25 0.39 0.23
We conducted a mixed model analysis for each interface to determine if learning or fatigue affected the user as they completed the numerical entry task. The analysis was performed by comparing the order that the tasks (T1 to T4) was given to that participant and the error rate. No significant difference was observed in performance during tasks T1 to T4. This conforms that there may be no learning or fatigue affected the participants during completing the numerical entry task.
4. RESULTS AND DISCUSSION A digit by digit analysis was conducted of each number entered to determine what type of errors were made. We have observed three categories of the errors as, interface (user interface- in this case designed software interface) errors, interaction error and numerical data entry errors. But only last category that is numerical data entry errors is considered for this study. The forty eight users entered 896 numbers, amounting to 3056 digits in total which includes digits added and omitted due to
Fig 3. Taxonomy of numerical data entry errors
170 The Analysis of the total incorrect digit entry for both input devices (mouse and keyboard) using paired ttest indicates that the mean errors for the native language numerical data entry by mouse (mean=0.40, sd=0.55) was significantly lower than that of using keyboard (mean=0.49, sd=0.68), t(96)=2.38, p=0.019. Also the time required to complete the task by mouse (Task-T1) (mean=169.10, sd=39.17) was significantly less compared to keyboard (TaskT2) (mean-177.92, sd=31.62), t(48)=-2.18, p=0.34. Figure 3 we report the different types of error that observed in our previous study experiment [5]. The classification and frequency of each input device incorrect digit entry errors is shown in figure 4.
Fig 4. The classification of numerical errors by input device type (Keyboard and Mouse) Discussion. There are trade-offs to consider when selecting a numerical entry input device, as validated by the data collected in our experiment. Our results showed that the mouse is 2.78% more accurate than the keyboard during numerical entry. Because difference between error rates of these two input devices. The factors such as recognized error rate or the speed of entry may become important factors in selecting the most appropriate input device. The results show a considerably higher number of errors by keyboard numerical data entry task in comparison to mouse numerical data entry task. It was also observed that users were slightly slower during numerical entry using keyboard. There is large variation in ‘Missing digit’ error rate which may due to user fixate twice as much on the keyboard during numerical entry [3]. It was noticed that in ‘Wrong digit’ category of errors, the participants enters more incorrect digits by mouse compared to keyboard (Fig. 1). This may be because many participants habituated of using calculators which is somewhat similar to keyboard number keys.
5. CONCLUSION AND FUTURE WORK In this paper we focused on analyzing the number of errors that occurred during interaction with numerical entry interface by mouse and keyboard. There are significant differences in accuracy and speed for two
input devices (mouse and keyboard), this may be due to less visual attention on interface during keyboard entry and errors like ‘Missing digit’ are occurred more. The result of this study may guide designers to design better system considering input devices like mouse and keyboard especially for rural or semiurban users. As future work we would like to repeat this study using other platforms of numerical entry like touchscreens, to find out error rates of touchscreens compare to other input devices like mouse and keyboard. REFERENCES [1] Barchard, K.A., and Pace, L.A., 2011. Preventing human error: The impact of data entry methods on data accuracy and statistical results. Computers in Human Behavior, Volume 27, Issue 5, Pages 1834-1839. [2] Eslambolchilar, P., Webster, J., and Niezen, G., 2013. The Evolution of Number Entry: a case study of the telephone. In Human-Computer Interaction–INTERACT 2013 (pp. 538-545). Springer Berlin Heidelberg. [3] Oladimeji, P., Thimbleby, H., and Cox, A., 2011. Number entry interface and their effects of error detection. INTERACT 2011. [4] Reason, J., 1990. Human Error. Cambridge Press. [5] Salve, S., Yammiyavar, P., 2013. Influence of local ‘language’ in data entry errors: A pilot study in the rural Indian setting. Human Computer Interactions (ICHCI), 2013 International Conference on , vol., no., pp.1,4. doi: 10.1109/ICHCI-IEEE.2013.6887812 [6] Singh, G., Findlater, L., Toyama, K., Helmer, S., Gandhi R., and Balakrishnan, R., 2009. Numeric paper forms for NGOs. In Proc. ICTD (2009).
