Task AnalysisTask Analysis

IACT 403 IACT 931 CSCI 324Human Computer Interface

Lecturer: Gene Awyzio

Room: 3.117Phone: 4221 4090Email: gene@uow.edu.au

OverviewOverview

What is task analysis?Task Analysis Methods

task decomposition knowledge based analysis entity-relationship techniques

Sources of InformationUses of Task Analysis.

What is Task Analysis?What is Task Analysis?

Methods of analysing people's jobs: what people do what things they work with what they must know

What is Task Analysis?What is Task Analysis?

Example: in order to clean the house

get the vacuum cleaner outfix the appropriate attachmentclean the roomswhen the dust bag gets full, empty itput the vacuum cleaner and tools away

Must know about:vacuum cleaners,their attachments, dust bags,

cupboards, rooms etc..

Approaches to task analysisApproaches to task analysis

Task decomposition splitting task into (ordered) subtasks

Knowledge based techniques what the user knows about the task and

how it is organised

Approaches to task analysisApproaches to task analysis

Entity-relation based analysis relationships between objects and

actions and the people who perform them

General method: observe

unstructured lists of words and actions organise

using notation or diagrams

Differences from other techniquesDifferences from other techniques

Systems analysis focus - system design

Task analysis focus - the user

Differences from other techniquesDifferences from other techniques

Cognitive models focus - internal mental state granularity - practiced `unit' task

Task analysis focus - external actions granularity - whole job However

much overlap in generaldifferences have exceptions

Task DecompositionTask Decomposition

Aims: describe the actions people do structure them within task subtask hierarchy describe order of subtasks

Focus on Hierarchical Task Analysis (HTA) It uses:

text and diagrams to show hierarchyplans to describe order.

Textual HTA descriptionTextual HTA description

Hierarchy description … 0. in order to clean the house

1. get the vacuum cleaner out2. fix the appropriate attachment3. clean the rooms

• 3.1. clean the hall• 3.2. clean the living rooms• 3.3. clean the bedrooms

4. empty the dust bag5. put vacuum cleaner and attachments away

Textual HTA descriptionTextual HTA description

... and plans Plan 0:

do 1 - 2 - 3 - 5 in that order.when the dust bag gets full do 4

Plan 3: do any of 3.1, 3.2 or 3.3 in any order depending on which rooms need cleaning

N.B. only the plans denote order.

Generating the hierarchyGenerating the hierarchy

get flat list of tasksgroup tasks into higher level tasksdecompose lowest level tasks furtherStopping rules How do we know when to

stop? Is “empty the dust bag" simple enough?

Purpose: expand only relevant tasksError cost: stop when P x C is smallMotor actions: lowest sensible level.

line under box means no further expansion.

Plans shown on diagram or written elsewhere.

Same information as: 0. make a cup of tea

1. boil water...

Diagrammatic HTADiagrammatic HTA

Refining the descriptionRefining the description

Given initial HTA (textual or diagram) How to check/improve it? Some heuristics:

paired actions• e.g., where is `turn on gas’

restructure• e.g., generate task `make pot’

balance• e.g., is `pour tea' simpler than making pot?

Generalise• e.g., make one cup or two... or more.

Refined HTA for making teaRefined HTA for making tea

Types of planTypes of plan

fixed sequence e.g., 1.1 - 1.2 -1.3

optional tasks e.g., if the pot is full 2

waiting for events e.g., when kettle boils 1.4

cycles plan 5 5.2 empty cups? for each guest 5.3

No

Yes

5.1

Types of planTypes of plan

time-sharinge.g., do 1; at the same time …

discretionarye.g., do any of 3.1, 3.2 or 3.3 in any order

mixturesmost plans involve several of the above.

Knowledge Based AnalysesKnowledge Based Analyses

Focus on: Objects - used in task Actions - performed Taxonomies represent levels of abstraction

Knowledge Based AnalysesKnowledge Based Analyses Example:

motor controls steering steering wheel,

indicators engine/speed

direct • ignition, accelerator,

foot brake gearing

• clutch, gear stick lights

external • headlights, hazard

lights internal

• courtesy light

wash/wipe wipers

• front wipers, rear wipers

washers • front washers, rear

washers heating

• temperature control, air direction, fan, rear screen heater

parking • hand brake, door lock

radio numerous others

TDH notationTDH notation

TDH - Task Description HierarchyThree types of branch point in taxonomy:

XORnormal taxonomyobject in one and only one branch

AND object must be in bothrepresents multiple classifications

OR weakest casecan be in one, many or none

TDH notationTDH notation

Example: wash/wipe AND

function XOR• wipe

– front wipers, rear wipers• wash

– front washers, rear washersposition XOR

• front– front wipers, front washers

• rear– rear wipers, rear washers.

Larger TDH exampleLarger TDH example kitchen item AND

shape XORdished

• mixing bowl, casserole, saucepan, soup bowl, glassflat

• plate, chopping board, frying pan function OR

preparation• mixing bowl, plate, chopping board

cooking• frying pan, casserole, saucepan

dining XOR• for food

– plate, soup bowl, casserole• for drink

– glass

More on TDHMore on TDH

Uniqueness rule: can the diagram distinguish all objects?

e.g., plate is:• kitchen item/shape( flat)/

function{preparation,dining(for food)}/• nothing else fits this description

More on TDHMore on TDH

Actions have taxonomy too: kitchen job OR

preparation• beating, mixing

cooking• frying, boiling, baking

dining• pouring, eating, drinking.

Abstraction and cutsAbstraction and cuts

After producing detailed taxonomy `cut' it to yield abstract view.

That is, ignore lower level nodes. e.g., cutting above shape and below

dining, plate becomes:kitchen item/function

{preparation,dining}/

Abstraction and cutsAbstraction and cuts

This is a term in Knowledge Representation Grammar

(KRG)These can be more complex:

`beating in a mixing bowl' becomeskitchen job(preparation)

•using a kitchen item/ function {preparation}/.

Entity-Relationship Based TechniquesEntity-Relationship Based Techniques

Emphasis on objects, actions and their relationships Similar to object-oriented analysis, but …

includes non-computer entitiesemphasises domain understanding not

implementation

Entity-Relationship Based TechniquesEntity-Relationship Based Techniques

Running example: `Vera's Veggies' - a market gardening firm

Owner/manager: Vera BradshawEmployees: Sam Gummage and Tony Peagreenvarious tools including a tractor `Fergie’two fields and a glasshousenew computer controlled irrigation system.

ObjectsObjects

Start with list of objects and classify them:Concrete objects:

simple things: spade, plough, glasshouseActors:

human actors: Vera, Sam, Tony, the customers what about the irrigation controller?

ObjectsObjects

Composite objects: sets: the team = { Vera, Sam, Tony } tuples: tractor may be < Fergie, plough >

To the objects add attributes: Object Pump3 simple - irrigation pump

Attributes:• status: on/off/faulty• capacity: 100 litres/minute

N.B. need not be computationally complete

ActionsActions

List actions and associate with each: agent - who performs the actions patient - which is changed by the action instrument - used to perform action

Examples: Sam (agent) planted (action) the leeks (patient) Tony dug the field with the spade (instrument)

ActionsActions

Note: implicit agents - read behind the words

`the field was ploughed' - by whom? indirect agency - the real agent?

`Vera programmed the controller to irrigate the field’ messages - a special sort of action

`Vera told Sam to …’ rôles - an agent acts in several rôles

Vera as worker or as manager.

E/R Example I - objects and actionsE/R Example I - objects and actions

Object Sam human actor Actions:

S1: drive tractor S2: dig the carrots

Object Vera human actor - the proprietor Actions: as worker

V1: plant marrow seed V2: program irrigation controller

Actions: as manager V3: tell Sam to dig the carrots

Object the men composite Comprises: {Sam, Tony}

E/R Example I - objects and actionsE/R Example I - objects and actions

Object glasshouse simple Attribute:

humidity: 0-100%

Object Irrigation Controller non-human actor Actions:

IC1: turn on Pump1IC2: turn on Pump2IC3: turn on Pump3

Object Marrow simple Actions:

M1: germinateM2: grow.

EventsEvents

Events are when something happens performance of action

`Sam dug the carrots’ spontaneous events

`the marrow seed germinated’`the humidity drops below 25%’

timed events`at midnight the controller ...'

RelationshipsRelationships

object - object social - Sam is subordinate to Vera spatial - pump 3 is in the glasshouse

action - object agent - (listed with object) patient and instrument

RelationshipsRelationships

actions and events temporal and causal

`Sam digs the carrots because Vera told him’

Temporal relations also use HTA or dialogue notations. show task sequence (normal HTA) show object lifecycle (see page 241).

E/R example II - events and relationsE/R example II - events and relations

Events Ev1: humidity drops below 25% Ev2: midnight

Relations: object - object location ( Pump3, glasshouse ) location ( Pump1, Parker's Patch )

E/R example II - events and relationsE/R example II - events and relations

Relations: action - object patient ( V3, Sam )

Vera tells Sam to dig patient ( S2, the carrots )

Sam digs the carrots … instrument ( S2, spade )

… with the spade

E/R example II - events and relationsE/R example II - events and relations

Relations: action - event before ( V1, M1 )

the marrow must be sown before it can germinate triggers ( Ev1, IC3 )

when humidity drops below 25%, the controller turns on pump 3

causes ( V2, IC1 ) the controller turns on the pump because Vera

programmed it.

Sources of InformationSources of Information

Documentation N.B. manuals say what is supposed to happen but, good for key words and prompting

interviewsObservation

formal/informal, laboratory/field (see Chapter 11)

Interviews the expert: manager or worker? (ask both!).

Early analysisEarly analysis

Extraction from transcripts list nouns (objects) and verbs(actions) beware technical language and context

`the rain poured’`I poured the tea’

Sorting and classifying grouping or arranging words on cards ranking objects/actions for task relevance (see

Ch. 11) use commercial outliner

Early analysisEarly analysis

Iterative process: data sources analysis

But costly, so use cheap sources where available.

Uses of Task Analysis IUses of Task Analysis I Manuals and

Documentation Procedural `how to do it'

manual from HTA description useful for extreme

novices or when domain too

difficult assumes all tasks

known

To make cups of tea

boil water - see page 2empty potmake pot - see page 3wait 4 or 5 minutespour tea - see page 4

-page 1-

Make pot of teaonce water has boiledwarm potput tea leaves in potpour in boiling water

-page 3-

Conceptual manualfrom knowledge or

entity/relation based analyses

good for open ended tasks

Example: tea making manual from HTA

Uses of Task Analysis IIUses of Task Analysis II

Requirements capture and systems design lifts focus from system to use suggests candidates for automation uncovers user's conceptual model

Uses of Task Analysis IIUses of Task Analysis II

Detailed interface design taxonomies suggest menu layout object/action lists suggest interface objects task frequency guides default choices existing task sequences guide dialogue design

NOTE. task analysis is never complete rigid task based design inflexible system

QuestionnairesQuestionnaires

Set of fixed questions given to users.Advantages:

quick and reaches large user group can be analysed more rigorously

Disadvantages less Flexible less probing

Need careful design what information is required? how are answers to be analysed?

Question TypesQuestion Types

Dichotomous Multiple Choice Multiple Response Open Ended Rank/Match Likert Semantic Differential Rating

Dichotomous Dichotomous

These are designed to be answered yes or no. The respondent should have the opportunity to answer "I don't know" or "I don't remember".

e.g. Do you use sampling in your production process? Yes

No

I don't know

Dichotomous Dichotomous

Advantages of dichotomous questions: easy to ask easy to understand quick to ask quick/easy to record easy to analyse

Disadvantage misunderstandings are possible as shades of

meaning are not incorporated to derive detailed information, a large number of such questions would need to be asked

Multiple Choice Multiple Choice

Multiple-choice questions permit only one answer. By convention, radio-buttons are used on computer screens where only one option is allowed. Respondent should also be able to answer "I don't know".

e.g. How would you rate the programme? Very PoorPoorFair

GoodVery Good

Multiple Choice Multiple Choice

Advantages of multiple-choice questions: allow many shades of meaning in answers give respondent freedom of choice easily recorded easy to analyse

Disadvantages difficult to ensure list of answers is complete difficult to phrase initially exclusive answers the list of answers should not be too long

Multiple Response Multiple Response

Multiple-response questions can have two or more answers.

By convention, check-boxes are used on computer screens where several options are allowed.

Respondent should also be able to answer "I don't know", "all" or "none".

Questions can have any number of responses or a fixed number as illustrated below. e.g. Which three of the

following best describe the show? Fun

nyInteresting

Educational

Diverting

Boring

Offensive

Multiple Response Multiple Response

Advantages of multiple-response questions allow many shades of meaning in answers give respondent freedom of choice easily recorded easy to analyse

Disadvantages difficult to ensure list of answers is complete difficult to phrase initially exclusive answers the list of answers should not be too long

Open Ended Open Ended

Designedto give respondent complete freedom of choice in answering and thus derive maximum information.

Often used where there is insufficient information available on a topic to permit complete lists of alternative answers.

Difficult to analyse with computers. e.g. How enjoyable did you find the presentation?

Open Ended Open Ended

Some structure can be added using the following techniques:

Word association Words are presented, and respondents note the first

word that comes to mind. e.g. What is the first word that comes to your mind

when you see the following? Your mother

Open Ended Open Ended

Sentence completion Incomplete sentences are presented, and respondents

complete them. e.g. When I choose a beer, the most important

consideration in my decision is

Open Ended Open Ended Advantages of open-ended questions

produce extensive information responses are free from bias of suggested answers many facets of respondent's behaviour revealed

Disadvantages can require slow verbatim recording interviewers tend to select information they believe significant lengthy interviews difficult to analyse many incoherent answers much irrelevant information received analysis requires groupings of answers - which loses much of the

shades of meanings

Rank/Match Rank/Match

e.g. Rank the following factors in order of importance when you buy a car

After sales support

Brand name

Value for money

Fuel economy

Appearance

Other (please specify)

Likert Likert

Likert questions are statements with which the respondent indicates the amount of agreement/ disagreement. e.g. Small shops generally give better service than

large onesStrongly agreeAgree

Neither

DisagreeStrongly disagree

Semantic Differential Semantic Differential

In this type of question a scale is inscribed between two bipolar words, and the respondent selects the point that represents the direction and intensity of his or her feelings.

e.g. How would you describe your views on shops opening on Sundays? Modern Old-fashioned

Question SequenceQuestion SequenceThere are a number of general rules which

should be considered when the order of the questions is arranged.

Those questions which are easy to answer should be put at the beginning, in order to give the respondent confidence in his ability to help you.

Those questions which are likely to interest the respondent should be early in the order.

The questions should be asked in a logical order.

Question SequenceQuestion SequenceFilter questions should follow each other

without being interrupted by other questions.Before a change of topic, introductory

phrases should be used to enable the respondent to make an easy transition.

Personal or emotional questions should be at the end.

More complicated questions should be at the end.

These guidelines tend to imply recommendation of a funnel design: