Name the steps in the Five-step procedure for data analysis

Name the steps in the Five-step
procedure for data analysis
Describe the step “Validation” in the
data analysis procedure.
Describe the step “Editing” in the data Describe the step “Coding” in the data
analysis procedure.
analysis procedure.
Describe the step “Data entry” in the
data analysis procedure.
Describe the step “Machine Cleaning”
in the data analysis procedure.
Describe “one-way frequency table”
and “cross tabulation”
Describe the difference between
“quantitative data” and “qualitative
data”
Describe the qualitative data
analyzing process 1
Describe the qualitative data
analyzing process 2
The process of ascertaining that data
was actually collected as specified.
Validation and editing, Coding, Data
Entry, Machine Cleaning, Tabulation
Coding refers to the process of
grouping and assigning numeric codes
to the responses. Different answers
get different numbers, which allows
you to group them easily.
Checking of completeness,
consistency and legibility of data
An error checking routine. A computer
program checks the data for logical
errors.
This refers to the process of
converting data from a form that
cannot be read by a computer to a
form that can.
Quantitative: Based on meanings derived from
numbers. The collection of data results in
numerical and standardized data and is analyzed
through diagrams and statistics.
One way: A table showing the number
of responses to each answer of a
question.
Qualitative: Based on meaning expressed
through words. Data-collection results in nonstandardized data requiring classification.
Analysis through the use of conceptualization.
1. Data reduction – summarizing and
simplifying the data collected or focusing
some part of the data.
2. Data display – organizing and assembling
your reduced or selected data into
diagrammatic or visual display.
3. Drawing and verifying conclusion – Find
patterns, regularity from the displayed
data.
Cross: Examination of the responses
to one question relative to responses
to one or more other question.
1. Categorization (Classify your data)
2.“Unitizing” (Attach units of data to relevant
category)
3.Recognizing relationships and developing
categories (Generate categories and
reorganizing your data according to them)
4. Develop and test theories to each conclusion
(When seeking patterns and relationships within
your data you will have to develop hypothesis in
order to test these)
Describe the qualitative data
analyzing process 1
Describe the qualitative data
analyzing process 2
Name four different types of analysis
you can use when handling qualitative
data
Describe the analytic tool Pattern
Matching, which can be used for
qualitative data analyses
Describe the analytic tool
Explanations building, which can be
used for qualitative data analyses
Describe the analytic tool Within-Case
analysis, which can be used for
qualitative data analyses
Describe the analytic tool Cross-case
Analysis, which can be used for
qualitative data analyses
4. Data reduction – summarizing and
simplifying the data collected or focusing
some part of the data.
5. Data display – organizing and assembling
your reduced or selected data into
diagrammatic or visual display.
6. Drawing and verifying conclusion – Find
patterns, regularity from the displayed
data.
Predict a pattern of outcomes based
on theoretical proposition in order to
explain what you expect to find. If the
pattern of your data matches with the
theoretical proposition, you will be
able to explain the data with this
theory.
Compare empirical results with frame
of reference: Theoretical proposition
1. Categorization (Classify your data)
2.“Unitizing” (Attach units of data to relevant category)
3.Recognizing relationships and developing categories
(Generate categories and reorganizing your data according
to them)
4. Develop and test theories to each conclusion (When
seeking patterns and relationships within your data you will
have to develop hypothesis in order to test these)
Pattern Matching, Explanation
building, Within-Case Analysis, Crosscase Analysis
1. Devise a theoretically based hypothesis, which you will
seek to test
2. Undertake data collection through an initial case study in
order to be able to compare the finding from this in relation
to the theory.
3. If necessary, change the theory to better fit the findings
from the first case study.
4. A new round of data collection in order to compare the
findings from this in relation to the new theory.
5. If necessary, same as nr 3
6. same as 4, New iteration.
Compare cases with each other
Ontology
Epistemology
Mode 1,2,3 – Knowledge creation
Basic research – purpose & context
Applied research – purpose&context
Research Process
Research&Methods
Assumptions about the grounds of
knowledge
About how one might begin to
understand the world and
communicate this knowledge to
fellow human beings
Purpose
- Expand knowledge of business and
management
- Results in universal principles
- Findings of general value
Context
- University studies
- Researcher ”decides”
1.
2.
3.
4.
5.
6.
7.
8.
Introduction
- General problem
- Research problem
- Research questions
Theory / Literatureoverview
Frameofreference
Method
Empiri / Data / Case
Analysis
Findings / Conclusions
Implications / Further Research
Assumptions which concerns the very
essence of the phenomena under
investigation
Mode 1
- Academic interest
- ”new knowledge”
Mode 2
- World ofpractice
- Practical relevant knowledge
Mode 3
- Broader ”societal” implications
- Knowledge of human relevance
Purpose
Understanding of particular
business/management
Results in solution to problem
New knowledge limited to particular problem
Findings of value to ”handle” particular problem
Context
-
Variety of organizations
Negotiate do bjectives
Research – ”to find things out” namely creating
trustworthy knowledge
Methods – How we do things.
Research Methods – How we do things in order
to create trustworthy knowledge.
- There should be a clear purpose & systematic
approach
Förklara begreppet:
Frame of Reference
Förklara kort NOMINAL scale of
measurement, vanligt användningsområde och
ge exempel
Vid ett research proposal ska man påvisa
följande:
1) Should-do-ability
2)Do-ability
3)Want-to-ability
Förklara dessa!
Förklara kort ORDINAL scale of
measurement, vanligt användningsområde och
ge exempel
Nämn några attributer som kännetecknar en
bra research topic
Förklara kort INTERVAL scale of
measurement, vanligt användningsområde och
ge exempel
Nämn TRE typology of theories och skillanden
mellan dem
Förklara kort RATIO scale of measurement,
vanligt användningsområde och ge exempel
Reasearch Philosphies:
Förklara:
Vad är Epistemology samt
Nämn FYRA scales of measurement
TRE sätt att se på detta
Beskrivning: Använder siffror för att identifiera
objekt, individer, händelser eller grupper.
Typisk användning: Klassificering, finna
likheter/olikheter
Typisk beskrivande statistik: Frekvensräkning,
procentsatser
Exempel: Kön (1) Man, (2) Kvinna
Beskrivning: Nummer ger information
om den relativa mängden av vissa egenskaper
Typisk användning: Ranking/betygsättning
Typisk beskrivande statistik: Median, typvärde,
Exempel: Ranka följande märken mellan 1-5 där
5 är det du tycker bäst om.
Beskrivning: Det man mäter tilldelas ett
numeriskt värde
Typisk användning: Skillnaden mellan två
mätvärden
Typisk beskrivande statistik: Typvärde, median
och aritmetiskt medelvärde
Exempel: Temperatur
Beskrivning: Det som mäts kan beskrivas med
ett kontinuerligt varierande numeriskt värde, och
det finns ett entydigt sätt att definiera ett
nollvärde.
Typisk användning: Bestämma kvoter
Typisk beskrivande statistik: harmoniskt och
geometriskt medelvärde
Exempel: Längd, vikt, befolkning
Epistemology är läran om kunskap
Positivism - Det finns en sanning! Bevisade
hypoteser är fakta
Postpositivism – nonfalsified hypteser är fakta
Realism -Direct Verkligheten är identiskt med
det våra sinnen visar oss
-Critical Våra sinnen ger en viss systematisk
relation till verkligheten men inte en strikt
avbildning
Interpretivism – Människor agerar utifrån deras
tolkning(interpretation)
Selection of theories and models best suited for
knowledge generation concerning formulated
research problem and research questions
Should-do-ability – Why should the study be
conducted?
Do-ability – Is it possible to conduct the study?
Want-to-do-ability – Researchers engagement to
the topic! (Is there interest and ability to
carry out the study?)
Genomförbart – genuint intresse till ämnet,
kunskaper, tid och resurser, tillgång till ”rätt”
data, länka samman problem till teori
Symmetry of potential outcomes – resultatet ska
ha liknande värde oavsätt vad man kommer fram
till
Viktigt att du uppfyller ”uppgiftens” kriterier ex.
tillämpningsbar.
Grand theories – leder till ett nytt sätt att tänka
och se på saker
Middle-range theoriesSubstantive theories – Increase restriction in
terms of general applicability
Nominal
Ordinal
Interval
Ratio
Reasearch Philosphies:
Förklara:
Vad är Ontology samt
Beskriv och förklara skillnaden mellan
”Inductive” och ”Deductive”
TVÅ sätt att se på detta
Reasearch Philosphies:
Förklara:
Vad är Axiology
Research design:
Benämn och förklara TRE olika mål man kan ha
som research purpose
Förklara skillnanden mellan kvantitativ och
kvalitativ
Beskriv”cross-sectional” och ”longitudinal
design”
Fortsätt meningen:
It is better to have an approximate answer to the
correct question than…
Beskriv för- och nackdelar med ”crosssectional” respektive ”longitudinal design”
Vad är bättre än att ha ett absolut svar till
”fel” fråga?
Nämn FEM olika typer av research Strategies
Inductive: Härleda slutsatser från empiriska
erfarenheter: Real life  Theory
Ontology: Är läran om det varande gällande hur
världen är beskaffad. (The nature of reality)
Deductive: Härlede slutsatser från antaganden
som utgår från bevis:
TheoryReal lifeImproved Theory
Objectivism: Opartiskt synsätt
Exploratory:
Discovery of Ideas and insights, understand what
variables are relevant
Descriptive:
Describe functions or characteristics, understand
relevant variables
Causal(Explanatory):
Determine Cause – Effect relations, understand
connection between variables
Cross-sectional design: Ger en bild av en
population vid EN tidpunkt eller under ett kort
tidsintervall
Longitudinal design: Upprepade observationer
av samma variabler under en längre tid
Subjectivism: Personligt färgat synsätt
Axiology är läran om värde eller kvalitet
(judgement of value)
Kvantitativ = hur mycket. Det handlar om
fakta(positivistic). Går att : Mäta, analysera,
statistiska tekniker
Kännetecken: ”en verklighet”, Generalisering av
kunskap, objektivt
Kvalitativ =värde och unikhet. Handlar om att
förstå, helhetssyn, sammanhang.
Kännetecken: ”fler verkligheter”, djup kunskap,
subjektivt
Fördelar cross sectional:
Representativ provtagning(sampling)
reaktioner på metodfel (response bias)
Fördelar longitudinal:
Upptäcker skillnader
Möjlighet att samla in stora mängder data
Precision(accuracy)
…an absolute answer to the wrong question
Fördelar för den ena är nackdelar för den andra
Experiment
Survey
Archival analysis
History
Case Study
Ett ungefärligt svar till “rätt” fråga!
Förklara Interpretive och functionalist samt
radical humanist och radical functionalist
paradigms
När ska man välja case study respektive
survey i förhållande till antalet variabler och
provgruppsstorlek
Radical states vill ändra på sakers tillstånd (state
of affairs)
De andra två är regulatory och vill ändra på
saker inom det nuvarande tillståndet.
Radical humanist och interpretive paradigms har
subjektiv syn, de andra två objektiv.
Många variabler och liten provgrupp = case study
Få variabler och stor provgrupp = survey
Define “a sample”
What are the pros of using
Non-Probability Samples?
Which are the different types of
Non-probability sampling techniques?
What are the pros of using
Probability Samples?
Which are the different types of
Probability sampling techniques?
Define “Probability sampling”
Define “Non-Probability sampling”
Explain “Simple Random Sampling”
Explain “Systematic Sampling”
Explain “Stratified Sampling”
1. The researcher can be sure of obtaining
information from a representative cross-section of
the population of interest.
2. Sampling error can be computed.
A subset from a larger population.
3. The survey results are projectable to the total
population.
- Simple random sampling
- Systematic sampling
- Proportional stratified sampling
- Disproportional stratified sampling
- Cluster sampling
1. Non-Probability samples cost less than probability
samples. This characteristic may have considerable
appeal in those situations where accuracy is not of
critical importance.
2. Non-Probability samples ordinarily can be
conducted more quickly than probability samples.
- Multi-stage sampling
A probability sample is a sample in
which every element of the
population has a known and equal
probability of being selected into the
sample. It is often associated with
survey and experimental research
strategies.
Is considered to be the purest form of
probability sampling. Means that you select the
sample at random from the sampling frame
used.
1. Assign a unique number to each case.
2. Select cases using random numbers until your
actual sample size is reached.
-
Convenience sampling
Judgment sampling
Purposive sampling
Quota sampling
Snowball sampling
The probability of each case being
selected from the total population is
not known and it is impossible to
answer research questions or to
address objectives that require you to
make statistical conclusions about the
characteristics of the population.
A modified simple random sampling where
probability samples are distinguished by the following
procedural steps:
Probability sampling in which the entire population is
numbered and elements are drawn at regular
intervals from the sampling frame.
1. The original population is divided into two or more
mutually exclusive subsets (e.g. male and female)
1. Number each of the cases with a unique number.
2. Select the first case using a random number.
3. Calculate the sampling fraction
(actual sampling size/total population).
4. Select subsequent cases using the sampling
fraction to determine the frequency.
2. Simple random samples of elements from the
different subsets are chosen independently from
each other.
Explain “Cluster Sampling”
Explain “ Multi-stage Sampling”*
Explain “Convenience Sampling”
Explain “Judgment Samples”
Explain “Quota Samples”
Explain “Snowball Samples”
Explain “Purposive Sampling”*
Why is “Probability sampling”
generally the favorable method to
use?
Define “Validity”
Name the four different types of
validity and explain them
Is a development of cluster sampling. It is
normally used to overcome problems
associated with a geographically spread
population when face-to-face contact is needed.
What you do is that you divide the groups into
smaller sub-groups and let these sub-groups
represent the population.
Is similar to stratified samples since you divide the
population into groups (clusters). The difference is that for
cluster sampling the sampling frame is the complete list of
clusters rather than a list of individual cases.
1. Choose the cluster grouping for your sampling frame.
2. Number each of the clusters with a unique number.
3. Select your sample using some form of random
sampling.
A non-probability sample in which the
A non-probability sample used
selection criteria is based on personal
primarily because they are easy to
judgment that the element is
collect. Means that you randomly
representative of the population
select cases that are easiest to obtain
studied.
for your sample.
Is commonly used when it is difficult to identify
members of the desired population. Selection
of additional respondents is based on referrals
from the initial respondents.
1. Make contact with one or two cases in the
population.
2. Ask these cases to identify further cases.
3. Ask these new cases to identify further cases.
Because of the need for:
-
Projectable totals
Low allowable errors
High population heterogeneity
Small non-sampling errors
High expected costs of errors
- Face Validity – Agreement that a question, scale or measure
appears logically to reflect accurately what it was intended to
measure.
- Content Validity – Does the content cover the
representative sample of the domain to be measured?
- Criterion Validity – Ability of a statistical test to make
accurate predictions.
- Construct Validity – Extent to which your measurement
questions actually measure the presence of those constructs
you intended them to measure.
Is entirely non-random since the
population is divided into subgroups
classified on the basis of the
researcher’s judgment. Normally used
for interview surveys. It is less costly
than probabilistic methods and can be
set up very quickly.
Enables you to use your judgment to
select cases that will best enable you
to answer your research questions
and to meet your objectives.
That we measure what we intend to
measure.
What is “External Validity”?
What is “Internal Validity”?
Explain the quality concepts
“Credibility”, “Transferability”,
“Dependability” and “Confirmability”
Define “Secondary data”
Define “Primary data”
What are the two main issues
concerning the use of secondary
data?
Explain the term “Observation”
What are the five dimensions of
Observational Approaches?
What are the advantages of
Observation Research?
What are the disadvantages of
Observational Research?
The extent to which findings can be
attributed to interventions
(ingripanden) rather than any flaws in
your research design.
The extent to which the research
results from a particular study are
generalisable to all relevant contexts.
Pieces of information that have been
gathered for other purpose and only
might be relevant to the problems at
hand.
- Credability – Correct identification and description
of the subject.
- Transferability – To what degree results can be
transferred to other context.
- Dependability – The reliability of a person to others
because of his/her integrity, truthfulness and
trustfulness. Changing social world can not give
reliability.
- Confirmability – Can results be confirmed by
another study (objectivity).
- Availability – finding secondary data you require is a
matter of detective work and you have to make a judgment
on the benefits of price and time of using secondary data.
- Is it appropriate for present needs? – The data is
collected for other purposes which may not match that of
your own research. It might not match your problem
definition or your population (unit of analysis). It is also
likely to be less current than data you collect yourself.
- Natural vs. Contrived situations
- Open vs. Disguised observation
- Structured vs. Unstructured
- Human vs. Machine observers
- Direct vs. Indirect observation
- Only behaviour and physical personal
characteristics can usually be examined. The
researcher does not learn about motives,
attitudes, intentions or feelings.
- Observation research can be time consuming
and costly if the observed behaviour occurs
rather infrequently.
Data collected from interviews,
questionnaires or observations to
solve the particular problem under
investigation.
The systematic observation,
recording, description, analysis and
interpretation of people’s behavior.
“What is done and why?”
- Observation Research provides the researcher the
opportunity to watch what people actually do rather
than relying on reports of what they do.
- This approach can avoid much of the biasing factors
caused by the interviewer and question structures
associated with the survey approach.
What is a “Participant observation”?
What types of data can be generated
by Participant observations?
Give examples of types of Machine
Observations
What is a “Structured Observation”?
What are the 3 major types of error in
Structured observations?
Explain the term “Subject error” in
structured observations
Explain the term “Time error” in
structured observations
Explain the term “Observer effect” in
structured observations
What different types of interview are
there? (Three ways of defining them)
Explain how the different types of
interviews can be used in exploratory,
descriptive and explanatory studies
- Primary Observations – You note down what
happened or what was said at the time.
- Secondary observations – Statements by
observers of what happened or was said.
- Experiential data – Those data on your
perceptions and feelings as you experience the
process you are researching.
It’s an observation where the researcher
attempts to participate fully in the lives and
activities of subjects and thus becomes a
member of their group, organization or
community. It is a qualitative method and deals
with the meaning attached to the actions.
This enables researchers to share their
experiences by not merely observing what is
happening but also feeling it.
- Traffic counters
Structured observation is systematic
and has a high level of predetermined - Scanner-based research
structure. Your concern would be in
measurement
quantifying behavior, i.e. its function - Physiological
EEG
is to tell you how often something
- Voice pitch analysis
happen rather than why it happens.
- Opinion and Behavior measurement
Errors that may occur when research
subjects are studied in situations that
are inconsistent with their normal
behavior patterns, leading to atypical
responses.
The impact of being observed on how
people act.
- Exploratory – In-depth interviews can be helpful to
find out “what is happening”. Semi-structured
interviews may also be used.
- Descriptive – Structured interviews can be used as a
means to identify general patterns.
- Explanatory – Semi-structured interviews may be
used in order to understand the relationships
between variables, such as those revealed from a
descriptive study. Structured interviews may also be
used.
- Subject error
- Time error
- Observer effect
Error, usually associated with
structured observations, where the
time at which the observation is being
conducted provides data that are
untypical of the time period in which
the events being studied would
normally occur.
- Structured – also referred to as standardized
or respondent interview.
- Semi-structured – also referred to as Nonstandardized or informant interview.
- In depth/Unstructured – also referred to as
Non-standardized or informant interview.
Define the term “Focus group”
How many people should usually
participate in a group interview?
How are participants for a group
interview normally chosen?
Why is it recommended to use
horizontal slices through an
organization when conducting group
interviews?
When conducting group interviews,
considering what issues may help
you?
What is your role/what are you
suppose to do as a moderator for a
Focus group?
What are the pros and cons of using
telephone interviews?
Explain the term “Structured
interview”
Explain the term “Semi-structured
interviews”
Explain the term “Unstructured
interviews”
Normally between 4-8 people
(can sometimes be as few as 2 and as many as
12)
Participants should be grouped so as not to
inhibit individual’s possible contributions. That
may be related to lack of trust or to perceptions
of status differences. Horizontal slices through
an organization will mean that within each
group, participants will have similar status and
work experience. In this way, group interview
can be conducted in a number of levels within an
organization.
A group interview where the topic is
defined clearly and precisely and
there is a focus on enabling and
recording interactive discussion
between participants.
Using Non-probability sampling and
for a specific purpose.
- To keep the group within the boundaries of
the topic discussed.
- To generate interest in the topic and
encourage discussion, whilst at the same time
not leading the group towards certain options.
Se Bilaga, Question 1.
Uses questionnaires based on a predetermined
and standardized or identical set of questions.
You read out each question and the record the
response, usually with pre-coded answers. They
are used to collect quantifiable data and
therefore referred to as “quantitative research
interviews”.
Pros:
- Make contact with people with whom it would
otherwise be impractical to face-to-face.
- Speed of data collection
- Lower costs
Cons:
- No personal contact
- Reduced reliability
- Difficult to develop complex questions
These are informal interviews. You would use
them to explore in-depth a general areas in
which you are interested. Therefore, refer to
these as “in-depth interviews”. There are no
predetermined questions to work through but
you have to have a clear idea of the aspects you
want to explore.
The researcher has a list of themes
and questions to be covered,
although these may vary from
interview to interview. The
organizational context and flow of the
conversation may change the order of
questions (or even delete or add
some).
Explain how personal contact can
influence an interview
What do you have to think about
when conducting an interview?
Explain interviewer and interviewee
bias
Interviewer bias is where the comments, the tone or nonverbal behavior of the interviewer creates bias in the way
interviewees respond to the questions. This may be where
you attempt to impose your own beliefs through the
questions you ask.
Interviewee bias may be caused by perceptions about the
interviewer. The person may try to answer in a way that
puts him in a “socially desirable” role or say positive things
about his organization just because.
An interview provides people with the
opportunity to reflect on events without
needing to write anything down. It also provides
the opportunity for the ones getting interviewed
to receive feedback and personal assurance
about how the info will be used. Personal
interview may receive higher response rate than
using questionnaires and the interviewer has
more control over who’s actually answering the
questions.
- The key to success is careful preparation.
- Have knowledge about the topic.
- Provide relevant info to participants before the
interview (for example a list of themes).
- Choose the location wisely!
- Wear clothes that will be generally accepted in the
setting.
- There is no second chance to make a first
impression, first minutes are important!
- How to record the interview.
- Give it enough time!
Bilaga
Question 1: When conducting group interviews, considering what issues may
help you?
- Where your research project occurs within an organisation the request
to participate in a group interview may be received by individuals as an
instruction rather than allowing them a choice about whether to take
part. This may lead to some level of non-attending or unreliable data.
- Where one or two people dominate the discussion, you should seek to
reduce their contributions carefully to bring others in.
- You have to ensure that participants understand each other’s
contributions and that you develop an accurate understanding of the
points being made.
- Consider the location and setting of the interview. Conduct the interview
in a natural setting, something like the manager’s office may lead to
participants not feeling relaxed. And also, if possible try to arrange the
seating in a circular fashion.
- You should plan to undertake three or four interviews with any one type
of participant. If you after this not is receiving any new information
means you have heard the full range of ideas.