Copyright 2006 John Wiley & Sons , Inc Chapter 9 – Task In The Organizational Context HCI: Developing Effective Organizational Information Systems Dov Te’eni Jane Carey Ping Zhang
Dec 21, 2015
Copyright 2006 John Wiley & Sons, Inc
Chapter 9 – Task In The Organizational Context
HCI: Developing Effective Organizational Information Systems
Dov Te’eniJane CareyPing Zhang
Copyright 2006 John Wiley & Sons, Inc
Learning Objectives
Know the attributes of the organizational tasks pertinent to HCI and how they provide the context of tool-level task.
Understand the need for task analysis in HCI. Understand how the work context affects HCI design
and differentiates between structured and non-structured tasks.
Understand how organizational-level tasks are decomposed hierarchically and mapped to tool-level task, which are described by the TSSL model.
Understand and apply a method for analyzing the HCI requirements for decision support.
Copyright 2006 John Wiley & Sons, Inc
Project Selection Project Planning
Requirements User-Needs TestDetermination
Context User TaskAnalysis Analysis Analysis
Evaluation FormativeMetrics Evaluation
Alternative Selection
Interface SpecificationFormativeEvaluation
Metaphor Design Media Design Dialogue Design Presentation Design
Coding Formative Summative
Evaluation Evaluation
Human Concerns:PhysicalCognitiveAffectiveUsefulness
Implementation
Project Selection and Planning
Analysis
Design
HC
I Prin
ciple
s an
d G
uid
elin
es
Re
qu
irem
en
tsDe
term
ina
tio
n
User Needs Test
Ta
skA
na
lysisC
on
textA
na
lysis
Copyright 2006 John Wiley & Sons, Inc
Introduction
Dowell and Long (1989) regard the workers and the information systems as “work systems” that affect some domain (e.g. customer services) through work aimed at achieving organizational goals by means of a set of tasks.
Eventually, task analysis; determines how task achieves the work must dictate not only which tasks achieve work but
also how they are allocated to the computer and the user.
Copyright 2006 John Wiley & Sons, Inc
Introduction
Topics covered include task, work and their relation to HCI that begins with; Examining the task, at both the
organizational level and the tool level, Concentrates on characteristics of
organizational tasks and finally, Examining the impact of work on user
requirements.
Copyright 2006 John Wiley & Sons, Inc
Introduction
The HCI developer needs to understand; How the work is structured and performed in order
to understand how workers divide and coordinate their work into specific tasks.
The dynamics of work patterns to recognize opportunities to support work over time.
The workplace, with its computers and other artefacts, does not always behave as expected.
People behave in unpredicted and unplanned ways.
Copyright 2006 John Wiley & Sons, Inc
Introduction
Task is central to HCI. The TSSL model describes HCI by representing the user’s view of the task and its human-computer implementation.
The organizational context affects HCI in two ways; Characteristics of the organizational tasks and
the workplace affect user behavior and user requirements
Specific organizational tasks map onto specific tool-level tasks.
Copyright 2006 John Wiley & Sons, Inc
2.0 Characteristics of Organizational Tasks
Selecting an office
View two sites
Select two sitesto view
Define viewattributes
• Computer displays list of sites, and user selects two sites• Each site is identified by name and picture. Users can select or deselect a site.• User points at a name or picture and then selects an unselected or deselects a selected. Sites are ordered by name.• Sites are displayed in a checkbox list and users point or tab to a checkbox and either click or press any key on the keyboard.
Figure 9.3 Hierarchical task decomposition, with both the organizational-level tasks and thetool-level task described by the TSSL levels of how to operate the computer.
Org
an
izatio
na
lL
ev
el ta
sk
To
ol
Le
ve
lta
sk
No
te : B
roke
n lin
es
ind
icate
Inco
mp
lete
de
scriptio
ns
TSSL
Copyright 2006 John Wiley & Sons, Inc
2.0 Characteristics Of Organizational Level Task Organizational-Level Task :-
A worker’s designed (intended and planned) effort to accomplish an organizational demand.
Structured Task :- Task that is well defined, has clear and explicit
goals, and can be accomplished by following predefined procedures.
Non-structured Task :- Task that is ill-defined, has ambiguous goals (if it
has any at all), and has no explicit procedures that can ensure successful completion of the task.
Copyright 2006 John Wiley & Sons, Inc
Table 9.1 summarizes the impact of task characteristics on the focus of HCI design. It emphasizes the differences in the impact of different types of task on the users and their performance because different types of task require different resources. The different impacts, in turn, determine the corresponding type of support.
2.0 Characteristics Of Organizational Level Task
Copyright 2006 John Wiley & Sons, Inc
Table 9.1 The importance of Task Structure to HCI
Typ
e and
Co
ntext o
f T
ask
Exa
mp
le
Main
Imp
lication
s
Structured, routine, uniform
, repetitive ta
sks
Intensive data
entry
Minim
al user e
ffort and ma
ximum
speed; m
inimal learn
ing time; errors due to inatte
ntion a
nd boredom
; frustration w
ith ill-fitted interfaces (e.g. no shortcuts); physiological
stress
Structured to sem
i-structured, routine but d
iverse
tasks
Word processing
Minim
al user e
ffort and ma
ximum
speed fo
r certain op
erations; piecem
eal b
ut efficient learning; recall and skill n
ecessary to operate; errors due to misapplication
of rules; fit to specific tasks an
d to specific work habits; lim
ited problem
solving and
calculations
Unstru
ctured tasks
Sup
ported decision ma
king
Com
prehension
; creativity and flexibility; complex problem
solving and planning; sim
ulations, calculatio
ns’ and m
ental models; e
rrors and biases in ju
dgments; fit to individual
style
Multiple ta
sks concu
rrent
Ope
rating seve
ral office systems
Managin
g attention to se
veral task; controlling states and opera
tions in several system
s; confusion and overload leading to errors and
frustration
Interrelated tasks
Project m
anagement
Com
prehension
; comp
lex problem
solving; m
emory; sim
ulation and plann
ing; flexibility
Copyright 2006 John Wiley & Sons, Inc
3.0 Work At The Office As Context – Tasks and their interrelations
Transaction Processing System (TPS) Organizational Processing System (OIS) Management Information System (MIS) Decision Support System (DSS)
3.1 C
haracteristics O
f D
ifferent T
ypes Of
Work
The
variety of tasks found in organizations is very large and best reflected in the various types of com
puter system
i.e.,
The
variety of tasks and of users is the basis for exam
ining the im
plications on the type of required computer support
Copyright 2006 John Wiley & Sons, Inc
3.1 Characteristics Of Different Types Of Work
Often very unstructured. Involves many interrelated tasks. Managerial work probably the most complex
that combines most types of tasks (Table 9.2).
Professional
Work :-
Copyright 2006 John Wiley & Sons, Inc
Table 9.2 Propositions about Managerial Work (Adapted from Kurke & Aldrich, 1983)
Managers perform a great quantity of work with little free time (34 different activities per day, 44 hours a week).
Managerial jobs are characterized by brevity, variety, and fragmentation (63 % of activities lasted less than 9 minutes; only 5 % lasted more than an hour).
Managers favor verbal over written contacts (desk work and tours take up only 29 % of their time and the rest is for phone calls and meetings).
Scheduled meetings consume more time than any other activity (4 meetings a day with half involving 3 people or less).
Managers link their organization with outsiders in a variety of ways (external contacts took 52 % of verbal contacts; internal contacts over a third).
Copyright 2006 John Wiley & Sons, Inc
With reference to Table 9.1, the description of work environment highlights the aspects of human information processing that are more likely to suffer from the limitations and biases mentioned previously; The typical manager is likely to suffer from information
overload. To function effectively (as monitor, disseminator,
spokesperson) information must flow to and from the manager quickly.
The cognitive limitations on memory and processing become a critical threshold.
Support is needed to: Reduce the amount and complexity of information
processing Increasing the capability of memory, processing,
attention and comprehension.
3.2 Information Processing Functions Supported By HCI
Copyright 2006 John Wiley & Sons, Inc
Table 9.1 The importance of Task Structure to HCI
Ch
aracteristics of M
anag
erial Wo
rk
Ch
aracteristics of C
lerical S
up
po
rt W
ork
User L
imitatio
ns
Informa
tion o
verload
Informa
tion o
verload
Mem
ory, atte
ntion, cognitive
processing
Concurrent tasks
overload
Concurrent task overloa
d
Cognitive m
anagem
ent
Com
plex problem
solving
Mem
ory, cognitive
processing
Use
of mu
ltiple m
edia for
comm
unica
tion
Percep
tual p
rocessing
Copyright 2006 John Wiley & Sons, Inc
3.3 Work ModelingT
he application of the ideas m
entioned earlier (the types of tasks, the interrelations am
ong tasks, and the w
ork habits of the w
orkers who
perform the
tasks) in a particular w
ork setting for a particular com
puter system
(or system
s) m
ay require a special analysis.
One approach that begins the developm
ent of IS by
lookingat the w
ork context is the “Contextual D
esign” (Beyer &
Holtzblatt, 1998).
Copyright 2006 John Wiley & Sons, Inc
1. Contextual inquiry2. Work modeling3. Consolidation4. Work design5. User environment design6. Test with customers7. Putting into practice
3.3 Work Modeling
Contextual has seven
parts :-
NO
TE
:-T
he first 3 p
arts areth
e main
con
cern
of
this ch
apter
Copyright 2006 John Wiley & Sons, Inc
3.3 Work Modeling
1. Contextual inquiry : Determines the stakeholders and attempts to
model how they work in practice. Using a variety of interview techniques, contextual inquiry represents the users’ work habits, motivation, needs, and desires.
Copyright 2006 John Wiley & Sons, Inc
3.3 Work Modeling
2. Work modeling : Represents the tasks performed by individual workers and
interacting workers. Modeling techniques resemble objects orientation tools and diagrams. One diagram represents the allocation of tasks to workers and another represents the interrelations between tasks. Additionally, each task is described individually (like a flowchart). Moreover, modeling also includes the interactions between workers, both the formal interactions around tasks and the informal interactions.
Work modeling goes further, however, by looking closely at the physical environment. It seeks to identify how workers use physical artifacts such as notes and desks to manage their work. (E.g. how a worker arranges his desk may indicate certain needs that would have to be satisfied in a paperless office.
Copyright 2006 John Wiley & Sons, Inc
3.3 Work Modeling
3. Consolidation : This is needed when (assumed in contextual
design) the design is done in teams. It combines the information gained from individual interviews to form a holistic view of work.
Copyright 2006 John Wiley & Sons, Inc
4.0 Decision Making As Organizational Task
4.1 T
he
Dec
ision
L
ife-C
ycle as a B
asis fo
r D
etermin
ing
R
equ
irem
ents
Helps to identify the opportunities for supporting the user (Te’eni, 1992).
Should not be confused with the HCI Development Life Cycle, which describes how to develop a computerized system.
Defines the main activities involved in a decision-making process and how they progress from one activity to another.
The life cycle can then be used to determine what resources are needed for each activity. what the user’s limitations and strengths are in performing the
activity what functionalities are needed to overcome the limitations and
extend the strengths. 5 main stages of the decision life cycle are shown in Figure 9.4.
Copyright 2006 John Wiley & Sons, Inc
Figure 9.4 Decision Life-CyclePre-decisional
activitiesInform
ation acquisition
Structuring and clarifying
information E
valuation of alternatives and choice
Post-decisional
activities
Copyright 2006 John Wiley & Sons, Inc
4.1 Decision Life-Cycle
1. Pre-decisional Activities : Include complex processes such as problem
detection, strategy formulation, and strategy selection.
2. Information Acquisition : Includes all forms of obtaining information
from external sources as well as retrieving information from one’s own memory.
Copyright 2006 John Wiley & Sons, Inc
4.1 Decision Life-Cycle
3. Structuring and Clarifying Information : Involves complex processes aimed at making
sense of the decision situation and the available information so as to be able to make a decision.
4. Evaluation of Alternatives and Choice : Is the most obvious part of decision making. Involves comparisons of alternatives with
some structure defined in the previous stage.
Copyright 2006 John Wiley & Sons, Inc
4.1 Decision Life-Cycle
5. Post-decisional Activities : Is often valuable to examine the activities after
the choice is made in order to identify further opportunities of supporting the user.
It is important to consider the affective aspects that follow from a decision i.e., the user’s confidence in and conformance with the decision made.
Learning from feedback about the decision and revising accordingly all or part of the earlier stages in the Decision Life Cycle.
Copyright 2006 John Wiley & Sons, Inc
1. Availability (Tversky & Kahneman, 1974). People tend to use information that readily available to them and ignore information that is less available, even if it is from reliable sources.
2. Confirming information ((Hogarth, 1987). People tend to seek and accept information that confirms rather than disconfirms their expectations.
3. Selective perceptions (Hogarth, 1987). The bias toward confirming information is the form of selective perception.
4.1 Decision Life-CycleInformation acquisition, both from
external and internal (m
emo
ry) sources, is subject to several biases. C
onsider the follow
ing 3 biases;
Copyright 2006 John Wiley & Sons, Inc
4.1 Decision Life-CycleT
he lim
itations and biases discussed in relation to each of the 5 stages in the D
ecision Life C
ycle im
ply several types of support for users engaged in decision m
aking. An
attempt to
organize the needs for supporting decision m
aking is show
n in T
able 9.4.
Copyright 2006 John Wiley & Sons, Inc
1. Working with the system, appropriate functionality can reduce the user’s effort in each step of the Decision Life Cycle (Figure 9.4).
2. It can reduce errors (e.g. in computation).3. It can affect the user’s selection of decision
strategy by making one strategy more attractive to the user than other strategies (Te’eni, 1989; Todd & Benbasat, 1992).
4.1 Decision Life-Cycle
How
system’s functionality affects user behavior;
Copyright 2006 John Wiley & Sons, Inc
Table 9.4 Common Needs of Decision Making (Silver, 1991)
Fuller exploration of alternatives. Earlier detection of problems. Coping with multiple objectives. Treat risk. Reduce cognitive biases. Creativity Communication Structure decision-making process Learning
Copyright 2006 John Wiley & Sons, Inc
4.2 Levels of Interaction in Decision Making (Figure 9.5)
An
exampl
e
of
the
appl
icat
ion
of
the
Hierarchi
cal
Task Decomposition.
Find the best location for a new
shop
List possible options
Predict no. ofcustom
ers
Estim
ate costs
View
the sites
Com
puter displaysall options for newshop
- Com
puter presentscustom
ers forecaststem
plate- U
ser inputsparam
eters- C
omputer
calculates anddisplays forecasts
- Com
puterpresents costtem
plate- U
ser inputsparam
eters- C
omputer
calculates anddisplays costs
- Com
puter displayslists of sites- U
ser selects one- C
omputer displays
the site- U
sers view it
To
ol
level
Decisio
n
level
Copyright 2006 John Wiley & Sons, Inc
1. Use of internal representations.2. Pursuit of goals.3. Chunking of information.
5.0 A Method For Task Analysis And Decision Support
5.1 An Overview
of the Method
Wane
Zachary
(1998) w
as one of the first to integrate design tools for D
SS
based on cognitive engineering. H
is m
ethodology assum
ed that hum
an decision m
aking depends on;
Copyright 2006 John Wiley & Sons, Inc
1. Working memory.2. Cognitive processing.3. Retrieval from long-term memory.4. Numerical operations.5. Projection in time and space.
5.0 A Method For Task Analysis And Decision Support
5.1 An Overview
of the Method
Furt
herm
ore, human decision m
aking is lim
ited by 5 factors:
Copyright 2006 John Wiley & Sons, Inc
5.0 A Method For Task Analysis And Decision Support
5.1 An Overview
of the Method
- Figure 9.7
describes a process of designing a D
SS
that is adapted from
Z
achary’s m
ethodology. It begins w
ith an analysis of the task and its constraints, continues w
ith a specification of the functionality required, and then proceeds to a design of the technology
- Z
achary’s m
ethodology assum
es that characteristics of decision m
aking lead to several recurring problem
s, w
hich can be m
inimized
by a system
atic process of analysis and design.
Copyright 2006 John Wiley & Sons, Inc
5.0 A Method For Task Analysis And Decision Support
Fi
gure 9.7 Analysis and Design of DSS
1. Situational and Functional Analysis (organizational context, work style, task attributes, objective and perceived
effectiveness, task dynamics, decision process, data and knowledge resources, problem representation, required judgments, errors)
2. Constraint Analysis (human limitations and biases, data constraints, communication failures,
stress)3. Propose New Functionality
(redefine elements of the decision process and the allocation of task to human and computer)
4. Design HCI, Models, Data, and Control (design detailed human-computer interfaces, computerized decision
techniques, data management, and control over the system operation throughout the human-computer interaction; develop prototype)
5. Reanalyze Work with New DSS and Modify (for objective and perceived effectiveness and for new constraints)
Copyright 2006 John Wiley & Sons, Inc
5.0 A Method For Task Analysis And Decision Support
The essence of this stage is to define the decision problem and decompose its goal into sub-goals.
Its main tool is a protocol for collecting and analyzing information
Its product is a Summary Tabulation of Aiding Requirements STAR Table (Figure 9.8)
The goal decomposition should; concentrate on concrete goals work top-down from the more general to the more specific
goals
5.2 Situational and F
unctional A
nalysis
Copyright 2006 John Wiley & Sons, Inc
Figure 9.8 STAR Table for Decision Support (adapted from Zachary, 1988)
Decision situation: (Name of decision situation) Task dynamics:
(state types – closed loop, iterative, unfolding, or single instance) Situational objective:
(highest level goal that the Decision Maker (DM) is trying to achieve) Value criteria:
(list of individual criteria by which alternative decisions are evaluated) Underlying process:
(a brief description of the observable process in which the decision situation is embedded) Information environment:
Inputs (list of available information items) Outputs (list of information items that are created) Parameters (list of information items that do not change value)
Intermediate reasoning / analysis steps: (list of intermediate reasoning steps or types of analysis that the DM typically applies in the
baseline process) Representation:
(a simple description including both the external world and the internal world) Judgment:
(required judgments, which are unstructured and intuitive activities)
Copyright 2006 John Wiley & Sons, Inc
5.0 A Method For Task Analysis And Decision Support
Identifies the decision-making needs that arise from the constraints on unaided or the baseline decision-making process.
The general direction is; to refer to common limitations or biases in in decision making to examine what is relevant to the particular decision situation
as characterized in the prior stage of situational and functional analysis.
5.3 Constraint
Analysis
Copyright 2006 John Wiley & Sons, Inc
5.0 A Method For Task Analysis And Decision Support
As stated earlier in Table 9.4 (Common Needs of Decision Making by Silver, 1991) in helping to overcome the limitations, Zachary (1988) too offers a similar list as shown in Table 9.5 (Common Limitations on Decision Making – Zachary, 1988) and relates them to the categories of the decision situation (Figure 9.8 - STAR Table for Decision Support – Zachary, 1988).
Tables 9.4 (Silver, 1991) and 9.5 (Zachary, 1988) are no more than checklists that need to be matched with the analysis documented in the STAR Table.
The product of this stage is a list of specific decision-making difficulties.
5.3 Constraint
Analysis
Copyright 2006 John Wiley & Sons, Inc
Table 9.5 Common Limitations on Decision Making (Zachary, 1988)
Inability to predict processes It is difficult to project processes due to limitations on working memory, attention to detail,
and numeric processing. Possible outcomes are projecting with errors, relying on suboptimal techniques, and ignoring information.
Difficult in combining attributes and objectives that are competing It is difficult to combine ill-defined objectives or conflicting objectives, due to limitations on
working memory and numeric processing. Possible outcomes are incorrect integrated criteria and omission of important attributes or objectives.
Inability to manage information Due to limitations on working memory and long-term memory and also limited attention.
Possible outcomes are errors in using the information and ignoring information intentionally and unintentionally.
Difficult in analyzing and reasoning Limitations on knowledge, time, and effort make it difficult to formulate an effective
decision strategy or to carry it out correctly. Difficulties in visualizing
It is difficult to visualize abstract manipulations and semantic data. It is easier to visualize concrete situations but difficult to manipulate them, due to limitations on working memory and cognitive processing. A major outcome is limiting the ability to manipulate information and therefore producing less effective solutions.
Copyright 2006 John Wiley & Sons, Inc
SUMMARY
Positions HCI within the organizational context by addressing task and work.
Differentiate between task at the organizational level and task at the tool level and position both on one task hierarchy, where tool-level tasks are the means for accomplishing the higher organizational-level task.
Looks at the characteristics of organizational tasks and work to help define user requirements by identifying needs or difficulties.
Eventually looks at methods for analyzing and decomposing tasks.
This
Chapter;
Copyright 2006 John Wiley & Sons, Inc
SUMMARY
WORK is the ensemble of interrelated TASKS and WORKERS who perform the task.
The general characteristics of a TASK, particularly the task’s degree of structure, determine the types of computer support that are most appropriate in order to boost performance, ease work, and generally satisfy the user.
The general characteristics of WORK, particularly the types of users and the interrelations between tasks performed by the same person or groups of workers, also dictate the limitations users face and the computer support they require.
The analysis of tasks in the work context they are performed is the basis for determining user requirements and designing HCI.