1、 ITU-T RECMNUF.910 95 W 48b2591 ObOO3bb T2 INTERNATIONAL TELECOMMUNICATION UNION ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU OPERATIONS AND QUALITY OF SERVICE HUMAN FACTORS F.91 O (02/95) PROCEDURES FOR DESIGNING, EVALUATING AND SELECTING SYMBOLS, PICTOGRAMS AND ICONS ITU-T Recommendation
2、F.91 O (Previously “CCITT Recommendation” ITU-T RECMN*F*SLO 95 4862591 O600367 919 FOREWORD The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff questions an
3、d issuing Recommen- dations on them with a view to standardizing telecommunications on a worldwide basis. The World Telecommunication Standardization Conference (WTSC), which meets every four years, establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendati
4、ons on these topics. The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in WTSC Resolution No. 1 (Helsinki, March 1-12, 1993). ITU-T Recommendation F.910 was prepared by ITU-T Study Group 1 (1993-1996) and was approved under the WTSC Resolution No. 1 pr
5、ocedure on the 21st February 1995. NOTE In this Recommendation, the expression ?Administration? is used for conciseness to indicate both a telecommunication administration and a recognized operating agency. O ITU 1995 All rights reserved. No part of this publication may be reproduced or utilized in
6、any form or by any means, electronic or mechanical, including photocopying and microfilm, without permission in writing from the ITU. ITU-T RECMN*F.SLO 75 = 4862571 Ob00368 855 9 CONTENTS 1 Background 2 Methodology 2.1 Determine the need for an international symbol, pictogram or icon 2.2 Design one
7、or more alternatives . 2.3 Evaluation procedures 2.4 Selection criteria 3 General parameters of experiments 4 Other considerations . Page 1 1 1 1 1 5 6 6 Recommendation F.910 (02/95) I ITU-T RECNN*F.SLO 95 W 4862593 ObOO3b9 791 = SUMMARY Various national, regional and international standards organiz
8、ations are responsible for the design, evaluation and standardization of symbols, pictograms and icons. In order to coordinate the work of these standards organizations, and in order to optimize the designs, it is desirable to follow a well-defined methodology. This Recommendation describes a framew
9、ork for a common methodology to be used by the ITU-T. To the extent that other standards organizations can work within this framework, the development of important designs should progress more rapidly. A uniform methodology will minimize duplication of effort and will maximize the relevance of colle
10、cted data. 11 Recommendation F.910 (02/95) ITU-T RECMN*F-SLO 95 48b259L Ob00370 403 Recommendation F.910 PROCEDURES FOR DESIGNING, EVALUATING AND SELECTING SYMBOLS, PICTOGRAMS AND ICONS (Genevu, 1994) 1 Background Various national (e.g. ANSI, CIAJ, DIN), international-regional (e.g. ETSI) and intema
11、tional-worldwide (e.g. ITU-T, ISO) standards organizations are responsible for the design, evaluation and standardization of symbols, pictograms and icons. In order to coordinate the work of these standards organizations, and in order to optimize the designs, it is desirable to follow a well-defined
12、 methodology. This Recommendation describes a framework for a common methodology to be used by the IT-T. To the extent that other standards organizations can work within this framework, the development of important designs should progress more rapidly. A uniform methodology will minimize duplication
13、 of effort and will maximize the relevance of collected datal). There are four high-level steps in the methodology for standardizing designs. First, a need must be determined. Second, designs must be created. Third, designs must be evaluated. Fourth, designs must be selected and approved. The steps
14、of this process are shown in Figure 1. The remainder of this Recommendation describes the methodology in detail. 2 Methodology 2.1 Two or more countries should express the need for standardization in order to make the proposed symbol, pictogram or icon of international concern. Determine the need fo
15、r an international symbol, pictogram or icon 2.2 2.2.1 Actual design process The ITU-T will not restrict suggested designs. The only requirement for suggested designs is that each must be usable in all appropriate forms (e.g. printed, engraved, embossed, screen-displayed). 2.2.2 Suggested designs wi
16、ll be published in ITU-T meeting reports and/or Special Rapporteurs reports which will include a request for alternative designs. Liaison statements will be sent to other relevant standards organizations. A reasonable time interval will be allowed for submissions before the evaluation procedures beg
17、in. The suggested reasonable time is a minimum of two months and a maximum of the interval from one Working Party meeting to the next. Design one or more alternatives Process for soliciting alternative designs from ITU-T Members 2.3 Evaluation procedures The various stages in the evaluation procedur
18、e are intended to measure initial (u priori) meaning and appropriateness, ease of learning, ease of remembering, and the probability of confusion with other designs. The order of procedures should be: a) b) explain, define and demonstrate functions for each design: evaluate contextually meaningful a
19、ssociations by using the matching task shown in Figure 4 and described below in 2.3.3 b); i) This Recommendation was developed with the help of ITU-T contributors who are also active in other standards organizations. Some procedures used by CIAJ and some used by ETSI (in its multiple index approach)
20、 have been adapted for this Recommendation. Recommendation F.910 (OU99 1 ITU-T RECMN*F.SLO 75 4862573 0600373 34T c) collect subjective data; d) if indicated by the subjective data, eliminate bad designs and create new candidate designs and then re- start the experiment; e) evaluate memory for meani
21、ng by a recall test; f) evaluate confusibility by a recognition-matching test. The evaluation should be a within-subjects design such that each person provides data in stages b), c), e) and 0. The recommended procedures can be performed in one session, or can be accomplished in a series of sessions
22、with the participants. If a small number of designs is under consideration, immediate memory for all of the designs may be perfect or nearly perfect, whereas delayed memory may show important differences in recall and/or recognition. It is advisable, therefore, for the evaluation coordinator to do a
23、 pilof experiment to determine a delay that wiI1 reveal differences in memory for the various designs. The evaluation coordinator should specify the procedural details before data are collected by the experimenters in various countries. 2.3.1 Explain, define and demonstrate the function of each desi
24、gn a) Evaluate preexisting associations - The matching task that is illustrated in Figure 4 and described in 2.3.3 b) is administered to determine the strength of pre-existing associations between a functions verbal description and a design. These data can be used to eliminate the worst designs and
25、to identify the most confusing designs. b) When possible, construct a prototype device with the test designs on it. Use this prototype to familiarize the subjects with the functions of each symbol, pictogram or icon. When construction of a prototype is not practical, the functions of the symbols tha
26、t will occur together should be explained in a scenario that corresponds to a scenario that would be typical when using a device. c) 2.3.2 Collect subjective data - After all the to-be-rated symbols have been described (explained), present the symbols one at a time to each subject to collect the sub
27、jective opinion and rating data. Figure 2 shows an example page for a questionnaire that can be used to collect these subjective data. In Figure 2, questions i) and ii) are to help identify those proposed designs that exist or are perceived to be too similar to designs that exist, whereas, questions
28、 iii) through vii) ask people to judge the designs a priori meaning, future memorability, and graphic qualities. If these data indicate high levels of confusion among designs or with existing designs, it may be necessary to intempt the evaluation to generate new designs as indicated in 2.3 d). If th
29、is situation occurs, the evaluation procedure should be re-started with a new set of participants. 2.3.3 Objective data a) Recall test The data from this test will measure the ability of subjects to recall the meaning of individually seen designs. These data are of particular relevance to situations
30、 where designs are experienced individually (e.g. on signs, sometimes on screens, etc.). The general procedure is to present a design (on-screen, projected, or on paper), then to allow time for the subject to write the meaning on an answer sheet (or, alternatively, to type the meaning if the designs
31、 are computer-presented). Whereas, it is desirable to promote the use of consistent verbal descriptions worldwide, it is recognized that this is difficult. Therefore, care must be taken to count as correct the various verbal expressions of meaning that are essentially the same. 2 Recommendation F.91
32、0 (02/95) ITU-T RECMN*F.S3D 95 i8b2591 0600372 286 v Determine Need I Design Evaluation need for intemational Not a ITU-T concern country only Two or more couniries Create symbol, pictogram or icon desigis. Publicize initiai designs. T=l Solicit alternative designs frUn ITU-T members. standards orqa
33、nizations with designs 6 be tested. Solicit additional designs. Specify evaluation procedures in detail. Publish for participating countries to follow. Determine preference by collecting subjective data Then explain or define each candidate designs function. Determine memory performance by collectin
34、g data (Recall memory test). Determine conf usability potential by collecting data r (Recognition matching test). Determine Riblish candidates for Recom- mendation by ITU-T and send liaison for comment to interested worldwide standards organizations. ITU-T TO 10361 0-94ld01 FIGURE 1E.910 Procedure f
35、or standardizing symbol, pictograms and icon designs Recommendation F.910 (02/95) 3 Function Description Hold Pressing the button holds a line to interrupt communication I. Have you evef seen this exact symbol (or pictogram or icon) before today? Yes. No. If yes, where? II. Have you ever seen a spbd
36、 that is wry similar to this me? Yes. No. If yes, where? Desim w - Please ans wer each of the next fne questbns by selecting the value on ihe five-poht rathg scale. You may also exphin yaw rating by wrihg a commeni in the space prvv the subjects comprehension of the procedures required to use the de
37、vice. While it may be too expensive to perform this optional final test, it may be important in some instances. I I Designs Function w - Thank you for your cooperation. T0103630-94ld03 FIGURE 3F.910 Recall memory test format Selection criteria a) Two or more design candidates per function When two o
38、r more designs are evaluated for a function, selection of the best design should follow these guidelines: - - - select the most preferred design (subjective rating data); select the design for which recall of its function was best (recall data); select the least confusable design (recognition data).
39、 If a particular design does not meet all three criteria, it may be necessary to create new design candidates. Alternatively. the design that best approximates these three criteria may be selected for standardization. The design that is determined to be the best of two or more design candidates shou
40、ld also meet the minimum criteria for acceptance listed in 2.4 b). Recommendation F.910 (OY95) 5 ITU-T RECNN*F.SLO 95 = YBb2591 Ob00375 T95 b) One design candidate perfunction When only one design is evaluated for a function, selection criteria are more difficult. In such situations we are not askin
41、g “which design candidate is best?, but rather “is this particular design good enough?. The minimum criteria for acceptance of a design are suggested to be as follows: - subjective ratings should be at least neutral (the mid-point on the five-point scale), and preferably should be positive; - recall
42、 of the function (meaning) for a design should be at least 50 per cent correct when averaged across subjects; - recognition matching should at least exceed chance. 3 General parameters of experiments a) There are two types of evaluation studies: In one type, only one design is evaluated for each fun
43、ction. Acceptance of a design under these conditions is based on comparison to arbitrary minimum preference and performance values. Suggested minimum criteria are given in 2.4 b). The second type of evaluation study includes two or more design candidates per function. In general, this is the preferr
44、ed type of study. Winning designs should still meet the minimum criteria given in 2.4 b). b) It may be desirable to test a whole set of designs that will be used together against an alternative whole set of designs. c) The minimum number of countries to provide data should be three, and these should
45、 represent as widely different cultures and/or markets as possible. A greater number of countries should participate whenever possible. d) Sampling Characteristics should be balanced so that the samples fully represent the user populations of interest. For example, it will be generally desirable to
46、test equal numbers of males and females, equal numbers of users from each decade of life (up to 19, 20-29, 30-39, etc.), and it may be important to sample evenly from various educational or work experience categories (e.g . clerical, technical, labourer, homemaker, etc.). For some designs it may be
47、important to include young children, elderly adults, or the sensory impaired. e) The presentation techniques for explaining, learning, rating, and memory testing can vary from one experiment to the next, but care must be taken to use techniques that control extraneous variables. Acceptable technique
48、s include paper-and-pencil, video presentation, and computer-based presentation. f) The number of people to be tested in each country is to be determined by the experimenters, but it is suggested that the minimum number be 20 per category. g) The statistical procedures and summaries are to be determ
49、ined by the experimenters. 4 Other considerations a) If a design is generally known and used in non-telecommunication contexts, some evaluation steps may be unnecessary to adopt the design for use in telecommunications. , b) If new designs are to be used in context with existing symbols, it is advisable to include the existing symbols in the recognition test to measure confusability. 6 Recommendation F.910 (02l95) I I Technical tem On-hook dialling Description Can dial without picking up the handset Design 3 W 5 6 7 Please write down any other observations that you ha= ab
