Upload
mohit-goel
View
224
Download
0
Embed Size (px)
Citation preview
7/26/2019 BR Solution Set 2013
1/25
BUSINESS RESEARCH MS 108
MBA II, JUNE 2013
Q1 STEPS IN RESEARCH PROCESS
Scientific research involves a systematic process that focuses on being objective and gathering a
multitude of information for analysis so that the researcher can come to a conclusion. This
process is used in all research and evaluation projects, regardless of the research method
(scientific method of inquiry, evaluation research, or action research). The process focuses on
testing hunches or ideas in a park and recreation setting through a systematic process. In this
process, the study is documented in such a ay that another individual can conduct the same
study again. This is referred to as replicating the study. !ny research done ithout documenting
the study so that others can revie the process and results is not an investigation using the
scientific research process. The scientific research process is a multiple"step process here the
steps are interlinked ith the other steps in the process. If changes are made in one step of theprocess, the researcher must revie all the other steps to ensure that the changes are reflected
throughout the process. #arks and recreation professionals are often involved in conducting
research or evaluation projects ithin the agency. These professionals need to understand the
eight steps of the research process as they apply to conducting a study.
Step 1$ Identift!eP"#$%e&
The first step in the process is to identify a problem or develop a research question. The research
problem may be something the agency identifies as a problem, some knoledge or information
that is needed by the agency, or the desire to identify a recreation trend nationally. %et us
consider a problem that the agency has identified is childhood obesity, hich is a local problem
and concern ithin the community. This serves as the focus of the study.
Step2'Re(ie)t!e*ite"+t"e
&o that the problem has been identified, the researcher must learn more about the topic under
investigation. To do this, the researcher must revie the literature related to the research
problem. This step provides foundational knoledge about the problem area. The revie of
literature also educates the researcher about hat studies have been conducted in the past, ho
these studies ere conducted, and the conclusions in the problem area. In the obesity study, the
revie of literature enables the programmer to discover horrifying statistics related to the long"term effects of childhood obesity in terms of health issues, death rates, and projected medical
costs. In addition, the programmer finds several articles and information from the 'enters for
isease 'ontrol and #revention that describe the benefits of alking *,*** steps a day. The
information discovered during this step helps the programmer fully understand the magnitude of
the problem, recogni+e the future consequences of obesity, and identify a strategy to combat
obesity (i.e., alking).
7/26/2019 BR Solution Set 2013
2/25
Step3' C%+"if t!e P"#$%e&
any times the initial problem identified in the first step of the process is too large or broad in
scope. In step - of the process, the researcher clarifies the problem and narros the scope of the
study. This can only be done after the literature has been revieed. The knoledge gained
through the revie of literature guides the researcher in clarifying and narroing the researchproject. In the eample, the programmer has identified childhood obesity as the problem and the
purpose of the study. This topic is very broad and could be studied based on genetics, family
environment, diet, eercise, self"confidence, leisure activities, or health issues. !ll of these areas
cannot be investigated in a single study/ therefore, the problem and purpose of the study must be
more clearly defined. The programmer has decided that the purpose of the study is to determine
if alking *,*** steps a day for three days a eek ill improve the individual0s health. This
purpose is more narroly focused and researchable than the original problem.
Step -' C%e+"% .efine Te"&/ +nd C#nept/
Terms and concepts are ords or phrases used in the purpose statement of the study or the
description of the study. These items need to be specifically defined as they apply to the study.
Terms or concepts often have different definitions depending on ho is reading the study. To
minimi+e confusion about hat the terms and phrases mean, the researcher must specifically
define them for the study. In the obesity study, the concept of 1individual0s health2 can be
defined in hundreds of ays, such as physical, mental, emotional, or spiritual health. 3or this
study, the individual0s health is defined as physical health. The concept of physical health may
also be defined and measured in many ays. In this case, the programmer decides to more
narroly define 1individual health2 to refer to the areas of eight, percentage of body fat, and
cholesterol. 4y defining the terms or concepts more narroly, the scope of the study is moremanageable for the programmer, making it easier to collect the necessary data for the study. This
also makes the concepts more understandable to the reader.
Step ' .efine t!e P#p%+ti#n
5esearch projects can focus on a specific group of people, facilities, park development,
employee evaluations, programs, financial status, marketing efforts, or the integration of
technology into the operations. 3or eample, if a researcher ants to eamine a specific group of
people in the community, the study could eamine a specific age group, males or females, people
living in a specific geographic area, or a specific ethnic group. %iterally thousands of options areavailable to the researcher to specifically identify the group to study. The research problem and
the purpose of the study assist the researcher in identifying the group to involve in the study. In
research terms, the group to involve in the study is alays called the population. efining the
population assists the researcher in several ays. 3irst, it narros the scope of the study from a
very large population to one that is manageable. Second, the population identifies the group that
the researcher0s efforts ill be focused on ithin the study. This helps ensure that the researcher
7/26/2019 BR Solution Set 2013
3/25
stays on the right path during the study. 3inally, by defining the population, the researcher
identifies the group that the results ill apply to at the conclusion of the study. 5esearcher has
identified the population of the study as children ages * to 6 years. This narroer population
makes the study more manageable in terms of time and resources.
Step ' .e(e%#p t!e In/t"&ent+ti#n P%+n
The plan for the study is referred to as the instrumentation plan. The instrumentation plan serves
as the road map for the entire study, specifying ho ill participate in the study/ ho, hen, and
here data ill be collected/ and the content of the program. In the obesity study, the researcher
has decided to have the children participate in a alking program for si months. The group of
participants is called the sample, hich is a smaller group selected from the population specified
for the study. The study cannot possibly include every * to 6"year"old child in the community,
so a smaller group is used to represent the population. The researcher develops the plan for the
alking program, indicating hat data ill be collected, hen and ho the data ill be
collected, ho ill collect the data, and ho the data ill be analy+ed. The instrumentation planspecifies all the steps that must be completed for the study. This ensures that the programmer has
carefully thought through all these decisions and that she provides a step"by"step plan to be
folloed in the study.
Step ' C#%%et .+t+
7nce the instrumentation plan is completed, the actual study begins ith the collection of data.
The collection of data is a critical step in providing the information needed to anser the
research question. 8very study includes the collection of some type of data9hether it is from
the literature or from subjects9to anser the research question. ata can be collected in theform of ords on a survey, ith a questionnaire, through observations, or from the literature. In
the obesity study, the programmers ill be collecting data on the defined variables$ eight,
percentage of body fat, cholesterol levels, and the number of days the person alked a total of
*,*** steps during the class. The researcher collects these data at the first session and at the last
session of the program. These to sets of data are necessary to determine the effect of the
alking program on eight, body fat, and cholesterol level. 7nce the data are collected on the
variables, the researcher is ready to move to the final step of the process, hich is the data
analysis.
Step 8' An+%4e t!e .+t+
!ll the time, effort, and resources dedicated to steps through : of the research process
culminate in this final step. The researcher finally has data to analy+e so that the research
question can be ansered. In the instrumentation plan, the researcher specified ho the data ill
be analy+ed. The researcher no analy+es the data according to the plan. The results of this
analysis are then revieed and summari+ed in a manner directly related to the research
questions. In the obesity study, the researcher compares the measurements of eight, percentage
7/26/2019 BR Solution Set 2013
4/25
of body fat, and cholesterol that ere taken at the first meeting of the subjects to the
measurements of the same variables at the final program session. These to sets of data ill be
analy+ed to determine if there as a difference beteen the first measurement and the second
measurement for each individual in the program. Then, the data ill be analy+ed to determine if
the differences are statistically significant. If the differences are statistically significant, the study
validates the theory that as the focus of the study. The results of the study also provide valuable
information about one strategy to combat childhood obesity in the community.
!s you have probably concluded, conducting studies using the eight steps of the scientific
research process requires you to dedicate time and effort to the planning process. 5esearcher
cannot conduct a study using the scientific research process hen time is limited or the study is
done at the last minute. 5esearchers ho do this conduct studies that result in either false
conclusions or conclusions that are not of any value to the organi+ation.
;6a) Qe/ti#nn+i"e' ! questionnaire is a researchinstrument consisting of a series
of questionsand other prompts for the purpose of gathering information from respondents.
!lthough they are often designed forstatisticalanalysis of the responses, this is not alays the
case. ;uestionnaires have advantages over some other types of surveysin that they are cheap, do
not require as much effort from the questioner as verbal or telephone surveys, and often have
standardi+ed ansers that make it simple to compile data. ! distinction can be made beteen
questionnaires ith questions that measure separate variables, and questionnaires ith questions
that are aggregated into either a scale or inde. ;uestionnaires ithin the former category are
commonly part of surveys, hereas questionnaires in the latter category are commonly part of
tests.
;uestionnaires ith questions that measure separate variables could for instance include
questions on$
preferences (e.g. political party)
behaviors (e.g. food consumption)
facts (e.g. gender)
;uestionnaires ith questions that are aggregated into either a scale or inde, include for
instance questions that measure$
latent traits (e.g. personality traits such as etroversion)
attitudes (e.g. toards immigration)
an inde (e.g. Social 8conomic Status)
http://en.wikipedia.org/wiki/Researchhttp://en.wikipedia.org/wiki/Questionhttp://en.wikipedia.org/wiki/Statisticshttp://en.wikipedia.org/wiki/Statisticshttp://en.wikipedia.org/wiki/Statistical_surveyhttp://en.wikipedia.org/wiki/Questionhttp://en.wikipedia.org/wiki/Statisticshttp://en.wikipedia.org/wiki/Statistical_surveyhttp://en.wikipedia.org/wiki/Research7/26/2019 BR Solution Set 2013
5/25
;6b)O$/e"(+ti#n'7bservation is ay of gathering data by atching behavior, events, or noting
physical characteristics in their natural setting. 7bservations can be overt (everyone knos they
are being observed) or covert (no one knos they are being observed and the observer is
concealed). The benefit of covert observation is that people are more likely to behave naturally if
they do not kno they are being observed. hen data collection from individuals is not a realistic option. If respondents are unilling or
unable to provide data through questionnaires or intervies, observation is a method that
requires little from the individuals from here data need to be collected.
;- T5PES O6 SAMP*IN7 .ESI7N
The sampling design can be broadly grouped on to basis vi+., representation and element
selection. 5epresentation refers to the selection of members on a probability or by other means.
8lement selection refers to the manner in hich the elements are selected individually anddirectly from the population. If each element is dran individually from the population at large,
it is an unrestricted sample. 5estricted sampling is here additional controls are imposed, in
other ords it covers all other forms of sampling. The classification of sampling design on the
basis of representation and element selection is shon belo$
E%e&ent Rep"e/ent+ti#n B+/i/
7/26/2019 BR Solution Set 2013
6/25
Se%eti#n P"#$+$i%it N#np"#$+$i%it
Un"e/t"ited Si&p%e "+nd#& C#n(eniene
5estricted
'omple random
Systematic
Stratified
'luster
ouble
#urposive
?udgment
;uota
Snoball
PROBABI*IT5 SAMP*IN7
#robability sampling is here each sampling unit in the defined target population has a knon
non+ero probability of being selected in the sample. The actual probability of selection for each
sampling unit may or may not be equal depending on the type of probability sampling design
used. Specific rules for selecting members from the operational population are made to ensure
unbiased selection of the sampling units and proper sample representation of the defined target
population. The results obtained by using probability"sampling designs can be generali+ed to the
target population ithin a specified margin of error. The different types of probability sampling
designs are discussed belo/
Un"e/t"ited #" Si&p%e R+nd#& S+&p%in9
In the unrestricted probability sampling design every element in the population has a knon,
equal non+ero chance of being selected as a subject. 3or eample, if * employees (n @ *) are to
be selected from -* employees (& @ -*), the researcher can rite the name of each employee in
a piece of paper and select them on a random basis. 8ach employee ill have an equal knon
probability of selection for a sample. The same is epressed in terms of the folloing formula/
#robability of selection @ Si+e of sample
""""""""""""""""""""""""""
Si+e of population
8ach employee ould have a *A-* or .--- chance of being randomly selected in a dran
sample. >hen the defined target population consists of a larger number of sampling units, a more
sophisticated method can be used to randomly dra the necessary sample. ! table of random
numbers can be used for this purpose. The table of random numbers contains a list of randomly
generated numbers. The numbers can be randomly generated through the computer programsalso. Bsing the random numbers the sample can be selected.
Re/t"ited #" C#&p%e: P"#$+$i%it S+&p%in9
!s an alternative to the simple random sampling design, several comple probability sampling
design can be used hich are more viable and effective. 8fficiency is improved because more
7/26/2019 BR Solution Set 2013
7/25
information can be obtained for a give sample si+e using some of the comple probability
sampling procedures than the simple random sampling design. The five most common comple
probability sampling designs vi+., systematic sampling, stratified random sampling, cluster
sampling, area sampling and double sampling are discussed belo$
S/te&+ti "+nd#& /+&p%in9
The systematic random sampling design is similar to simple random sampling but requires that
the defined target population should be ordered in some ay. It involves draing every nth
element in the population starting ith a randomly chosen element beteen and n. In other
ords individual sampling units are selected according their position using a skip interval. The
skip interval is determined by dividing the sample si+e into population si+e. 3or e.g. if the
researcher ants a sample of ** to be dran from a defined target population of ***, the skip
interval ould be *(***A**). 7nce the skip interval is calculated, the researcher ould
randomly select a starting point and take every * thuntil the entire target population is proceeded
thorough. The steps to be folloed in a systematic sampling method are enumerated belo/ Total number of elements in the population should be identified
The sampling ratio is to be calculated ( n @ total population si+e divided by si+e of the
desired sample)
The random start should be identified
! sample can be dran by choosing every nth entry
To important considerations in using the systematic random sampling are/
It is important that the natural order of the defined target population list be unrelated to
the characteristic being studied. Skip interval should not correspond to the systematic change in the target population.
St"+tified R+nd#& S+&p%in9
Stratified random sampling requires the separation of defined target population into different
groups called strata and the selection of sample from each stratum. Stratified random sampling is
very useful hen the divisions of target population are skeed or hen etremes are present in
the probability distribution of the target population elements of interest. The goal in stratification
is to minimi+e the variability ithin each stratum and maimi+e the difference beteen strata.
The ideal stratification ould be based on the primary variable under study. 5esearchers oftenhave several important variables about hich they ant to dra conclusion. ! reasonable
approach is to identify some basis for stratification that correlates ell ith other major
variables. It might be a single variable like age, income etc or a compound variable like on the
basis of income and gender.
7/26/2019 BR Solution Set 2013
8/25
Stratification leads to segmenting the population into smaller, more homogeneous sets of
elements. In order to ensure that the sample maintains the required precision in terms of
representing the total population, representative samples must be dran from each of the smaller
population groups.
There are three reasons as to hy a researcher chooses a stratified random sample/
To increase the sample0s statistical efficiency
To provide adequate data for analy+ing various sub population
To enable different research methods and procedures to be used in different strata.
raing a stratified random sampling involves the folloing steps/
etermine the variables to use for stratification
Select proportionate or disproportionate stratification
ivide the target population into homogeneous subgroups or strata
Select random samples from each stratum
'ombine the samples from each stratum into a single sample of the target population.
There are to common methods for deriving samples from the strata vi+., proportionate and
disproportionate. In proportionate stratified sampling, each stratum is properly represented so
the sample dran from it is proportionate to the stratum0s share of the total population. The
larger strata are sampled more because they make up a larger percentage of the target population.
This approach is more popular than any other stratified sampling procedures due to the folloing
reasons/
It has higher statistical efficiency than the simple random sample It is much easier to carry out than other stratifying methods
It provides a self"eighting sample ie the population mean or proportion can be
estimated simply by calculating the mean or proportion of all sample cases.
In disproportionate stratified sampling, the sample si+e selected from each stratum is
independent of that stratum0s proportion of the total defined target population. This approach is
used hen stratification of the target population produces sample si+es that contradict their
relative importance to the study.
!n alternative of disproportionate stratified method is optimal allocation. In this method,
consideration is given to the relative si+e of the stratum as ell as the variability ithin the
stratum to determine the necessary sample si+e of each stratum. The logic underlying the optimal
allocation is that the greater the homogeneity of the prospective sampling units ithin a
particular stratum, the feer the units that ould have to be selected to estimate the true
population parameter accurately for that subgroup. This method is also opted for in situation
7/26/2019 BR Solution Set 2013
9/25
7/26/2019 BR Solution Set 2013
10/25
In stratified sampling the population is divided into a fe subgroups, each ith many
elements in it and the subgroups are selected according to some criterion that is related to
the variables under the study. In cluster sampling the population is divided into many
subgroups each ith a fe elements in it. The subgroups are selected according to some
criterion of ease or availability in data collection.
Stratified sampling should secure homogeneity ithin the subgroups and heterogeneity
beteen subgroups. 'luster sampling tries to secure heterogeneity ithin subgroups and
homogeneity beteen subgroups.
The elements are chosen randomly ithin each subgroup in stratified sampling. In cluster
sampling the subgroups are randomly chosen and each and every element of the subgroup
is studied in"depth.
.#$%e /+&p%in9
This is also called sequential or multiphase sampling. ouble sampling is opted hen furtherinformation is needed from a subset of group from hich some information has already been
collected for the same study. It is called as double sampling because initially a sample is used in
the study to collect some preliminary information of interest and later a subsample of this
primary sample is used to eamine the matter in more detail The process includes collecting data
from a sample using a previously defined technique. 4ased on this information, a sub sample is
selected for further study. It is more convenient and economical to collect some information by
sampling and then use this information as the basis for selecting a sub sample for further study.
NONPROBABI*IT5 SAMP*IN7
In non"probability sampling method, the elements in the population do not have any probabilities
attached to being chosen as sample subjects. This means that the findings of the study cannot be
generali+ed to the population.
7/26/2019 BR Solution Set 2013
11/25
eploratory phase of a research project and it is the best ay of getting some basic information
quickly and efficiently. The assumptions is that the target population is homogeneous and the
individuals selected as samples are similar to the overall defined target population ith regard to
the characteristics being studied.
7/26/2019 BR Solution Set 2013
12/25
assurance that prespecified subgroups of the defined target population are represented on
pertinent sampling factors that are determined by the researcher. It ensures that certain groups are
adequately represented in the study though the assignment of the quota.
Sn#)$+%% S+&p%in9
Snoball sampling is a non"probability sampling method in hich sets of respondents are
chosen ho helps the researcher to identify additional respondents to be included in the study.
This method of sampling is also called as referral sampling because one respondent refers other
potential respondents. Snoball sampling is typically used in research situations here the
defined target population is very small and unique and compiling a complete list of sampling
units is a nearly impossible task. >hile the traditional probability and other non"probability
sampling methods ould normally require an etreme search effort to qualify a sufficient
number of prospective respondents, the snoball method ould yield better result at a much
loer cost. The researcher has to identify and intervie one qualified respondent and then solicit
his help to identify other respondents ith similar characteristics.
;Ca)S"(e E""#"/'8rrors may occur at any stage during the collection and processing of
survey data, hether it is a census or a sample survey. There are to main sources of survey
error$ Sampling error (errors associated directly ith the sample design and estimation methods
used) and non"sampling error (a blanket term used to cover all other errors). &on"sampling errors
are usually sub"divided as follos$
'overage errors, hich are mainly associated ith the sampling frame, such as missing
units, inclusion of units not in the population of interest, and duplication.
5esponse errors, hich are caused by problems related to the ay questions ere
phrased, the order in hich the questions ere asked, or respondentsD reporting errors(also referred to as measurement error if possible errors made by the intervieer are
included in this category).
&on"response errors, hich are due to respondents either not providing information or
providing incorrect information. &on"response increases the likelihood of bias in the
survey estimates. It also reduces the effective sample si+e, thereby increasing theobserved sampling error.
7/26/2019 BR Solution Set 2013
13/25
!ll of these sources may contribute to either, or both, of the to types of survey error. These are
bias, or systematic error, and variance, or random error.
Sampling error is not an error in the sense of a mistake having been made in conducting the
survey. 5ather it indicates the degree of uncertainty about the DtrueD value based on information
obtained from the number of people that ere surveyed.
It is reasonably straightforard for knoledgeable, eperienced survey"taking organi+ations to
control sampling error through the use of suitable sampling methods and to estimate its impact
using information from the sample design and the achieved sample. !ny statement about
sampling errors, namely variance, standard error, margin of sampling error or coefficient of
variation, can only be made if the survey data come from a probability sample.
The non"sampling errors, especially potential biases, are the most difficult to detect, to control
and to measure, and require careful planning, training and testing.
;Cb) Re%i+$i%it'Internal consistency reliability is used to assess the reliability of a summated
scale here several items are summed to form a total score. In internal consistency reliability
estimation a single measurement instrument is administered to a group of people on one occasion
to estimate reliability. In effect the reliability of the instrument is judged by estimating ho ell
the items that reflect the same construct yield similar results. >e are looking at ho consistent
the results are for different items for the same construct ithin the measure.
There are a ide variety of internal consistency measures that can be used$
i< A(e"+9e Inte"ite& C#""e%+ti#n
The average inter"item correlation uses all of the items on the instrument that are designed to
measure the same construct. The correlation beteen each pair of items is computed first. 3oreample, if e have si items e ill have G different item pairings (i.e., G correlations). The
average inter item correlation is simply the average or mean of all these correlations.
ii< A(e"+9e Ite& t#t+% C#""e%+ti#n
This approach also uses the inter"item correlations. In addition, a total score for the si items is
computed and use that as a seventh variable in the analysis. The figure shos the si item"to"
total correlations at the bottom of the correlation matri.
iii< Sp%itH+%f Re%i+$i%it
In split"half reliability all items that purport to measure the same construct into to sets are
randomly divided. The entire instrument is administered to a sample of people and the total scorefor each randomly divided half is calculated.
7/26/2019 BR Solution Set 2013
14/25
i(< C"#n$+!=/ A%p!+ >?@
'ronbach0s alpha is a statistic. It is generally used as a measure of internal consistency or
reliability of a psychometric instrument. In other ords, it measures ho ell a set of variablesor items measures a single, one"dimensional latent aspect of individuals.
The value of alpha (H) may lie beteen negative infinity and .
7/26/2019 BR Solution Set 2013
15/25
description of several commonly used statistical tools, decision support models, and optimi+ation
routines
;uantitative arket 5esearch ecision Support Tools
Statistical ethods
ultiple 5egression " This statistical procedure is used to estimate the equation ith the
best fit for eplaining ho the value of a dependent variable changes as the values of a
number of independent variables shift. ! simple market research eample is theestimation ofthe best fitfor advertising by looking at ho sales revenue (the dependent variable) changes
in relation to ependitures on advertising, placement of ads, and timing of ads.
3actor !nalysis " This statistical method is used to determine hich are the strongest
underlying dimensionsof a larger set of variables that are inter"correlated. >here manyvariables are correlated, factor analysis identifies hich relations are strongest. ! Bsing factor
analysis, a market researcher ho ants to kno hat combination of variables or factors are
most appealing to a particular type of consumer can use factor analysis to reduce the datadon to a fe variables are most appealing to consumers.
'onditions for a 3actor !nalysis 8ercise$ It requires metric data. This means that the data should be either interval or ratio scale
in nature.
The si+e of the sample respondents should be at least four to five times more than the
number of variables. Initial set of variables should be highly correlated. ! correlation matri of the variables
could be computed and tested for its statistical significance. The test is carried out by
using a 4arttlet test of sphericity, hich takes the determinant of the correlation matri
into consideration.
Laiser"eyer"7lkinis carried out before performing factor analysis, it takes a valuebeteen * and .The L7 statistics compares the magnitude of observed correlation
coefficients ith magnitudes of partial correlation coefficients.
iscriminant !nalysis " This statistical technique is used to for classification of people,
products, or other tangibles into to or more categories. arket research can make use of
discriminant analyses in a number of ays. 7ne simple eample is to distinguishhatadvertising channelsare most effective for different types of products.
ajor assumptions of iscriminant !nalysis are$
the observations are a random sample.
each predictor variable is normally distributed/
each of the allocations for the dependent categories in the initial classiM cation are correctly classiM ed/
there must be at least to groups or categories, ith each case belonging to only one
group so that the groups are mutually eclusive and collectively ehaustive (all cases
can be placed in a group)/
each group or category must be ell deM ned, clearly differentiated from any other
group(s) and natural. #utting a median split on an attitude scale is not a natural ay to
http://marketresearch.about.com/od/market.research.segmentation/a/Discrete-Choice-Versus-Conjoint.htmhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Structural-Equation-Modeling.htmhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Structural-Equation-Modeling.htmhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Structural-Equation-Modeling.htmhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Structural-Equation-Modeling.htmhttp://marketresearch.about.com/od/market.research.brand.equity/a/Customer-Experience-Dynamics.htmhttp://marketresearch.about.com/od/market.research.brand.equity/a/Customer-Experience-Dynamics.htmhttp://marketresearch.about.com/od/market-research-quantitative/ss/Step-by-Step-to-Product-Launch-Market-Research-With-Bayesian-Networks_3.htmhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Structural-Equation-Modeling.htmhttp://marketresearch.about.com/od/market.research.segmentation/a/Discrete-Choice-Versus-Conjoint.htmhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Structural-Equation-Modeling.htmhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Structural-Equation-Modeling.htmhttp://marketresearch.about.com/od/market.research.brand.equity/a/Customer-Experience-Dynamics.htmhttp://marketresearch.about.com/od/market.research.brand.equity/a/Customer-Experience-Dynamics.htmhttp://marketresearch.about.com/od/market-research-quantitative/ss/Step-by-Step-to-Product-Launch-Market-Research-With-Bayesian-Networks_3.htmhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Structural-Equation-Modeling.htm7/26/2019 BR Solution Set 2013
16/25
form groups. #artitioning quantitative variables is only justiM able if there are easily
identiM able gaps at the points of division/
for instance, three groups taking three available levels of amounts of housing loan/
the groups or categories should be deM ned before collecting the data/
the attribute(s) used to separate the groups should discriminate quite clearly beteen
the groups so that group or category overlap is clearly non"eistent or minimal/
group si+es of the dependent should not be grossly different and should be at least M ve
times the number of independent variables.
iscriminat !nalysis is used hen$
the dependent is categorical ith the predictor IK0s at interval level such as age,
income,
attitudes, perceptions, and years of education, although dummy variables can be used
as predictors as in multiple regression. %ogistic regression IK0s can be of any level of
measurement.
there are more than to K categories, unlike logistic regression, hich is limited to a
dichotomous dependent variable.
'luster !nalysis " This statistical procedure is used to separate objectsinto a specific
number of groups that are mutually eclusive but that are also relatively homogeneous inconstitution. This process is similar to hat occurs in market segmentationhere the market
researcher is interested in the similarities that facilitate grouping consumers into segments and
is also interested in the attributes that make the market segments distinct.3or eample, Jouneed to identify people ith similar patterns of past purchases so that you can tailor your
marketing strategies.
The objective of cluster analysis is to identify groups of object that are very similar ith regardto their price consciousness and brand loyalty and assign them into clusters. !fter having decided
on the clustering variables (brand loyalty and price consciousness), e need to decide on the
clustering procedure to form our groups of objects. This step is crucial for the analysis, as
different procedures require different decisions prior to analysis.
'onjoint !nalysis " This statistical method is used to unpack thepreferences of consumersith
regard to different marketing offers. To dimensions are of interest to the market researcher
inconjoint analysis$ () The inferred utility functions of each attribute, and (6) the relative
importance of the preferred attributes to the consumers. 3or eample a computer may be
described in terms of attributes such as processor type, hard disk si+e and amount of memory.8ach of these attributes is broken don into levels " for instance levels of the attribute for
memory si+e might be =4, 6=4, -=4 and C=4.
These attributes and levels can be used to define different products or product profiles. The first
stage in conjoint analysis is to create a set of product profiles hich customers or respondents are
http://marketresearch.about.com/od/market-research-quantitative/a/Surveys-Research-Bayesian-Networks.htmhttp://marketresearch.about.com/sitesearch.htm?q=market+research+market+segmentation&SUName=marketresearchhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Conjoint-Analysis.htmhttp://marketresearch.about.com/od/market.research.techniques/ss/Creating-And-Using-Profiles-For-Conjoint-Analysis.htmhttp://marketresearch.about.com/od/market.research.techniques/ss/Creating-And-Using-Profiles-For-Conjoint-Analysis.htmhttp://marketresearch.about.com/od/market-research-quantitative/a/Surveys-Research-Bayesian-Networks.htmhttp://marketresearch.about.com/sitesearch.htm?q=market+research+market+segmentation&SUName=marketresearchhttp://marketresearch.about.com/od/market.research.techniques/a/Using-Conjoint-Analysis.htmhttp://marketresearch.about.com/od/market.research.techniques/ss/Creating-And-Using-Profiles-For-Conjoint-Analysis.htm7/26/2019 BR Solution Set 2013
17/25
then asked to compare and choose from. 7bviously, the number of potential profiles increases
rapidly for every ne attribute added, so there are techniques to simplify both the number of
profiles to be tested and the ay in hich preferences are discovered.ifferent type of conjoint
analysis (eg choice based conjoint, full"profile conjoint, or adaptive conjoint analysis) havedifferent approaches to coping ith the balance beteen attribute number and amount of data
that needs to be collected.
4y analysing hich items are chosen or preferred from the product profiles offered to the
customer it is possible to ork out statistically both hat is driving the preference from the
attributes and levels shon, but more importantly, give an implicit numerical valuation for each
attribute and level " knon as utilities or part"orths and importance scores.
The result is a detailed picture of ho customers make decisions and a set of data that can be
used to build market modelshich can predict market share in ne market conditions and test
the impact of product or service changes on the market to see here and ho you can gain the
greatest improvements over your competitors. 4asic assumptions of conjoint analysis are$
The product is a bundle of attributes
Btility of a product is a simple function of the utilities of the attributes
Btility predicts behavior (i.e., purchases)
Q S+%e >/#i+% /iene/@
In the social sciences, /+%in9is the process of measuringor ordering entities ith respect to
quantitative attributes or traits. 3or eample, a scaling technique might involve estimating
individualsD levels of etraversion, or the perceived quality of products. 'ertain methods of
scaling permit estimation of magnitudes on a continuum, hile other methods provide only for
relative ordering of the entities.
See level of measurementfor an account of qualitatively different kinds of measurement scales.
http://www.dobney.com/Conjoint/conjoint_flavours.htmhttp://www.dobney.com/Conjoint/conjoint_flavours.htmhttp://www.dobney.com/Conjoint/conjoint_flavours.htmhttp://www.dobney.com/Conjoint/ModelDemo.htmhttp://www.dobney.com/Conjoint/ModelDemo.htmhttp://en.wikipedia.org/wiki/Measurementhttp://en.wikipedia.org/wiki/Quantitativehttp://en.wikipedia.org/wiki/Continuum_(mathematics)http://en.wikipedia.org/wiki/Level_of_measurementhttp://www.dobney.com/Conjoint/conjoint_flavours.htmhttp://www.dobney.com/Conjoint/conjoint_flavours.htmhttp://www.dobney.com/Conjoint/ModelDemo.htmhttp://en.wikipedia.org/wiki/Measurementhttp://en.wikipedia.org/wiki/Quantitativehttp://en.wikipedia.org/wiki/Continuum_(mathematics)http://en.wikipedia.org/wiki/Level_of_measurement7/26/2019 BR Solution Set 2013
18/25
C#&p+"+ti(e +nd n#n #&p+"+ti(e /+%in9
>ith comparative scaling, the items are directly compared ith each other (eample $ o you
prefer #epsi or 'okeN). In noncomparative scalingeach item is scaled independently of the
others (eample $
7/26/2019 BR Solution Set 2013
19/25
C#&p+"+ti(e /+%in9 te!nie/
P+i")i/e #&p+"i/#n /+%e" a respondent is presented ith to items at a time and
asked to select one (eample$ o you prefer #epsi or 'okeN). This is an ordinal level
technique hen a measurement model is not applied.
R+n#"de" /+%e " a respondent is presented ith several items simultaneously and
asked to rank them (eample$ 5ate the folloing advertisements from to *.). This is anordinal level technique.
C#n/t+nt /& /+%e" a respondent is given a constant sum of money, script, credits, or
points and asked to allocate these to various items (eample $ If you had ** Jen to spend
on food products, ho much ould you spend on product !, on product 4, on product ',
etc.). This is an ordinal level technique.
QS#"t /+%e" Bp to C* items are sorted into groups based a rank"order procedure.
7tt&+n /+%e" This is a procedure to determine hether a set of items can be rank"
ordered on a unidimensional scale. It utili+es the intensity structure among several
indicators of a given variable. Statements are listed in order of importance. The rating is
scaled by summing all responses until the first negative response in the list. The =uttmanscale is related to 5asch measurement/ specifically, 5asch models bring the =uttman
approach ithin a probabilistic frameork.
N#n#&p+"+ti(e /+%in9 te!nie/
C#ntin#/ "+tin9 /+%e(also called the graphic rating scale) " respondents rate items byplacing a mark on a line. The line is usually labeled at each end. There are sometimes a
series of numbers, called scale points, (say, from +ero to **) under the line. Scoring and
codification is difficult.
*ie"t /+%e " 5espondents are asked to indicate the amount of agreement or
disagreement (from strongly agree to strongly disagree) on a five" or seven"point scale.
The same format is used for multiple questions.
P!"+/e #&p%eti#n /+%e/" 5espondents are asked to complete a phrase on an "point
response scale in hich * represents the absence of the theoretical construct and *
represents the theori+ed maimum amount of the construct being measured. The samebasic format is used for multiple questions.
Se&+nti diffe"enti+% /+%e" 5espondents are asked to rate on a : point scale an item on
various attributes. 8ach attribute requires a scale ith bipolar terminal labels.
St+pe% /+%e" This is a unipolar ten"point rating scale. It ranges from OG to "G and has no
neutral +ero point.
http://en.wikipedia.org/wiki/Pairwise_comparisonhttp://en.wikipedia.org/w/index.php?title=Rank-order_scale&action=edit&redlink=1http://en.wikipedia.org/w/index.php?title=Constant_sum_scale&action=edit&redlink=1http://en.wikipedia.org/w/index.php?title=Q-Sort_scale&action=edit&redlink=1http://en.wikipedia.org/wiki/Guttman_scalehttp://en.wikipedia.org/w/index.php?title=Continuous_rating_scale&action=edit&redlink=1http://en.wikipedia.org/wiki/Likert_scalehttp://en.wikipedia.org/wiki/Phrase_completionshttp://en.wikipedia.org/wiki/Semantic_differential_scalehttp://en.wikipedia.org/w/index.php?title=Stapel_scale&action=edit&redlink=1http://en.wikipedia.org/wiki/Pairwise_comparisonhttp://en.wikipedia.org/w/index.php?title=Rank-order_scale&action=edit&redlink=1http://en.wikipedia.org/w/index.php?title=Constant_sum_scale&action=edit&redlink=1http://en.wikipedia.org/w/index.php?title=Q-Sort_scale&action=edit&redlink=1http://en.wikipedia.org/wiki/Guttman_scalehttp://en.wikipedia.org/w/index.php?title=Continuous_rating_scale&action=edit&redlink=1http://en.wikipedia.org/wiki/Likert_scalehttp://en.wikipedia.org/wiki/Phrase_completionshttp://en.wikipedia.org/wiki/Semantic_differential_scalehttp://en.wikipedia.org/w/index.php?title=Stapel_scale&action=edit&redlink=17/26/2019 BR Solution Set 2013
20/25
T!"/t#ne /+%e " This is a scaling technique that incorporates the intensity structure
among indicators.
The folloing table classifies the various simple data types, associated distributions,
permissible operations, etc. 5egardless of the logical possible values, all of these data types
are generally coded using real numbers, because the theory of random variablesoften
eplicitly assumes that they hold real numbers.
.+t+ TpeP#//i$%e
(+%e/
E:+&p%e
/+9e
*e(e%
#f
&e+/
"e&en
t
.i/t"i$ti#n
S+%e
#f
"e%+ti(
e
diffe"
ene/
Pe"&i//i$%e
/t+ti/ti/
Re9"e//i#n
+n+%/i/
$in+"
*,
(arbitrary
labels)
binary
outcome
(EyesAnoE,
EtrueAfalse
E,
EsuccessAf
ailureE,
etc.) nomin
al
scale
4ernoulli
incom
parabl
e
mode, 'hi"
squared
logistic,prob
it
+te9#"i+
%
, 6, ..., L
(arbitrary
labels)
categorica
l outcome
(specific
blood
type,politi
cal party,
ord,
etc.)
categorical
multinomial
logit,multino
mial probit
#"din+% integerorre
al
number(arb
itrary
relative
score,
significan
t only for
ordinal
scale
categoricalNN relativ
e
compa
ordinal
regression(or
dered
logit,ordered
http://en.wikipedia.org/wiki/Thurstone_scalehttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Random_variablehttp://en.wikipedia.org/wiki/Random_variablehttp://en.wikipedia.org/wiki/Binary_variablehttp://en.wikipedia.org/wiki/Nominal_scalehttp://en.wikipedia.org/wiki/Nominal_scalehttp://en.wikipedia.org/wiki/Nominal_scalehttp://en.wikipedia.org/wiki/Bernoulli_distributionhttp://en.wikipedia.org/wiki/Comparabilityhttp://en.wikipedia.org/wiki/Comparabilityhttp://en.wikipedia.org/wiki/Comparabilityhttp://en.wikipedia.org/wiki/Mode_(statistics)http://en.wikipedia.org/wiki/Chi-squared_testhttp://en.wikipedia.org/wiki/Chi-squared_testhttp://en.wikipedia.org/wiki/Logistic_regressionhttp://en.wikipedia.org/wiki/Probit_regressionhttp://en.wikipedia.org/wiki/Probit_regressionhttp://en.wikipedia.org/wiki/Probit_regressionhttp://en.wikipedia.org/wiki/Categorical_variablehttp://en.wikipedia.org/wiki/Categorical_variablehttp://en.wikipedia.org/wiki/Blood_typehttp://en.wikipedia.org/wiki/Blood_typehttp://en.wikipedia.org/wiki/Political_partyhttp://en.wikipedia.org/wiki/Political_partyhttp://en.wikipedia.org/wiki/Categorical_distributionhttp://en.wikipedia.org/wiki/Multinomial_logithttp://en.wikipedia.org/wiki/Multinomial_logithttp://en.wikipedia.org/wiki/Multinomial_probithttp://en.wikipedia.org/wiki/Multinomial_probithttp://en.wikipedia.org/wiki/Ordinal_variablehttp://en.wikipedia.org/wiki/Integerhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Ordinal_scalehttp://en.wikipedia.org/wiki/Ordinal_scalehttp://en.wikipedia.org/wiki/Categorical_distributionhttp://en.wikipedia.org/wiki/Ordinal_regressionhttp://en.wikipedia.org/wiki/Ordinal_regressionhttp://en.wikipedia.org/wiki/Ordered_logithttp://en.wikipedia.org/wiki/Ordered_logithttp://en.wikipedia.org/wiki/Ordered_logithttp://en.wikipedia.org/wiki/Ordered_probithttp://en.wikipedia.org/wiki/Thurstone_scalehttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Random_variablehttp://en.wikipedia.org/wiki/Binary_variablehttp://en.wikipedia.org/wiki/Nominal_scalehttp://en.wikipedia.org/wiki/Nominal_scalehttp://en.wikipedia.org/wiki/Nominal_scalehttp://en.wikipedia.org/wiki/Bernoulli_distributionhttp://en.wikipedia.org/wiki/Comparabilityhttp://en.wikipedia.org/wiki/Comparabilityhttp://en.wikipedia.org/wiki/Comparabilityhttp://en.wikipedia.org/wiki/Mode_(statistics)http://en.wikipedia.org/wiki/Chi-squared_testhttp://en.wikipedia.org/wiki/Chi-squared_testhttp://en.wikipedia.org/wiki/Logistic_regressionhttp://en.wikipedia.org/wiki/Probit_regressionhttp://en.wikipedia.org/wiki/Probit_regressionhttp://en.wikipedia.org/wiki/Categorical_variablehttp://en.wikipedia.org/wiki/Categorical_variablehttp://en.wikipedia.org/wiki/Blood_typehttp://en.wikipedia.org/wiki/Blood_typehttp://en.wikipedia.org/wiki/Political_partyhttp://en.wikipedia.org/wiki/Political_partyhttp://en.wikipedia.org/wiki/Categorical_distributionhttp://en.wikipedia.org/wiki/Multinomial_logithttp://en.wikipedia.org/wiki/Multinomial_logithttp://en.wikipedia.org/wiki/Multinomial_probithttp://en.wikipedia.org/wiki/Multinomial_probithttp://en.wikipedia.org/wiki/Ordinal_variablehttp://en.wikipedia.org/wiki/Integerhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Ordinal_scalehttp://en.wikipedia.org/wiki/Ordinal_scalehttp://en.wikipedia.org/wiki/Categorical_distributionhttp://en.wikipedia.org/wiki/Ordinal_regressionhttp://en.wikipedia.org/wiki/Ordinal_regressionhttp://en.wikipedia.org/wiki/Ordered_logithttp://en.wikipedia.org/wiki/Ordered_logithttp://en.wikipedia.org/wiki/Ordered_logithttp://en.wikipedia.org/wiki/Ordered_probit7/26/2019 BR Solution Set 2013
21/25
scale)creating a
rankingrison probit)
$in#&i+% *, , ..., &
number of
successes
(e.g. yes
votes) out
ofNpossi
ble
interva
l
scaleNN
binomial,beta
"binomial,
etc.
additiv
eNN
mean,median,
mode,standard
deviation,corr
elation
binomial
regression(lo
gistic,probit)
#nt
nonnegativ
eintegers(*
, , ...)
number of
items(telephon
e calls,
people,
molecules
, births,
deaths,
etc.) in
given
intervalAar
eaAvolum
e
ratio
scale
#oisson,negat
ive binomial,
etc.
multip
licativ
e
!ll statistics
permitted for
interval scales
plus the
folloing$geo
metric
mean,harmoni
c
mean,coefficie
nt of variation
#oisson,
negative
binomial
regression
"e+%
(+%ed+d
diti(e
real
number
temperatu
re,
relative
distance,l
ocation
parameter
, etc. (or
approim
ately,
anything
not
varying
interva
l scale
normal, etc.
(usually
symmetric
about
themean)
additiv
e
mean,median,
mode,standard
deviation,corr
elation
standardline
ar regression
http://en.wikipedia.org/wiki/Ordered_probithttp://en.wikipedia.org/wiki/Binomial_variablehttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Binomial_distributionhttp://en.wikipedia.org/wiki/Beta-binomial_distributionhttp://en.wikipedia.org/wiki/Beta-binomial_distributionhttp://en.wikipedia.org/wiki/Beta-binomial_distributionhttp://en.wikipedia.org/wiki/Meanhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Mode_(statistics)http://en.wikipedia.org/wiki/Standard_deviationhttp://en.wikipedia.org/wiki/Standard_deviationhttp://en.wikipedia.org/wiki/Correlationhttp://en.wikipedia.org/wiki/Correlationhttp://en.wikipedia.org/wiki/Binomial_regressionhttp://en.wikipedia.org/wiki/Binomial_regressionhttp://en.wikipedia.org/wiki/Logistic_regressionhttp://en.wikipedia.org/wiki/Logistic_regressionhttp://en.wikipedia.org/wiki/Probit_regressionhttp://en.wikipedia.org/wiki/Count_variablehttp://en.wikipedia.org/wiki/Integerhttp://en.wikipedia.org/wiki/Integerhttp://en.wikipedia.org/wiki/Ratio_scalehttp://en.wikipedia.org/wiki/Ratio_scalehttp://en.wikipedia.org/wiki/Poisson_distributionhttp://en.wikipedia.org/wiki/Negative_binomial_distributionhttp://en.wikipedia.org/wiki/Negative_binomial_distributionhttp://en.wikipedia.org/wiki/Geometric_meanhttp://en.wikipedia.org/wiki/Geometric_meanhttp://en.wikipedia.org/wiki/Geometric_meanhttp://en.wikipedia.org/wiki/Harmonic_meanhttp://en.wikipedia.org/wiki/Harmonic_meanhttp://en.wikipedia.org/wiki/Harmonic_meanhttp://en.wikipedia.org/wiki/Harmonic_meanhttp://en.wikipedia.org/wiki/Coefficient_of_variationhttp://en.wikipedia.org/wiki/Coefficient_of_variationhttp://en.wikipedia.org/wiki/Coefficient_of_variationhttp://en.wikipedia.org/wiki/Poisson_regressionhttp://en.wikipedia.org/wiki/Real-valuedhttp://en.wikipedia.org/wiki/Real-valuedhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Location_parameterhttp://en.wikipedia.org/wiki/Location_parameterhttp://en.wikipedia.org/wiki/Location_parameterhttp://en.wikipedia.org/wiki/Location_parameterhttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Normal_distributionhttp://en.wikipedia.org/wiki/Meanhttp://en.wikipedia.org/wiki/Meanhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Mode_(statistics)http://en.wikipedia.org/wiki/Standard_deviationhttp://en.wikipedia.org/wiki/Standard_deviationhttp://en.wikipedia.org/wiki/Correlationhttp://en.wikipedia.org/wiki/Correlationhttp://en.wikipedia.org/wiki/Linear_regressionhttp://en.wikipedia.org/wiki/Linear_regressionhttp://en.wikipedia.org/wiki/Ordered_probithttp://en.wikipedia.org/wiki/Ordered_probithttp://en.wikipedia.org/wiki/Binomial_variablehttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Binomial_distributionhttp://en.wikipedia.org/wiki/Beta-binomial_distributionhttp://en.wikipedia.org/wiki/Beta-binomial_distributionhttp://en.wikipedia.org/wiki/Meanhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Mode_(statistics)http://en.wikipedia.org/wiki/Standard_deviationhttp://en.wikipedia.org/wiki/Standard_deviationhttp://en.wikipedia.org/wiki/Correlationhttp://en.wikipedia.org/wiki/Correlationhttp://en.wikipedia.org/wiki/Binomial_regressionhttp://en.wikipedia.org/wiki/Binomial_regressionhttp://en.wikipedia.org/wiki/Logistic_regressionhttp://en.wikipedia.org/wiki/Logistic_regressionhttp://en.wikipedia.org/wiki/Probit_regressionhttp://en.wikipedia.org/wiki/Count_variablehttp://en.wikipedia.org/wiki/Integerhttp://en.wikipedia.org/wiki/Ratio_scalehttp://en.wikipedia.org/wiki/Ratio_scalehttp://en.wikipedia.org/wiki/Poisson_distributionhttp://en.wikipedia.org/wiki/Negative_binomial_distributionhttp://en.wikipedia.org/wiki/Negative_binomial_distributionhttp://en.wikipedia.org/wiki/Geometric_meanhttp://en.wikipedia.org/wiki/Geometric_meanhttp://en.wikipedia.org/wiki/Geometric_meanhttp://en.wikipedia.org/wiki/Harmonic_meanhttp://en.wikipedia.org/wiki/Harmonic_meanhttp://en.wikipedia.org/wiki/Harmonic_meanhttp://en.wikipedia.org/wiki/Coefficient_of_variationhttp://en.wikipedia.org/wiki/Coefficient_of_variationhttp://en.wikipedia.org/wiki/Poisson_regressionhttp://en.wikipedia.org/wiki/Real-valuedhttp://en.wikipedia.org/wiki/Real-valuedhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Location_parameterhttp://en.wikipedia.org/wiki/Location_parameterhttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Interval_scalehttp://en.wikipedia.org/wiki/Normal_distributionhttp://en.wikipedia.org/wiki/Meanhttp://en.wikipedia.org/wiki/Meanhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Mode_(statistics)http://en.wikipedia.org/wiki/Standard_deviationhttp://en.wikipedia.org/wiki/Standard_deviationhttp://en.wikipedia.org/wiki/Correlationhttp://en.wikipedia.org/wiki/Correlationhttp://en.wikipedia.org/wiki/Linear_regressionhttp://en.wikipedia.org/wiki/Linear_regression7/26/2019 BR Solution Set 2013
22/25
over a
large
scale)
"e+%
(+%ed&
%tip%i+ti(
e
positive rea
l number
price,
income,
si+e,scale
parameter
, etc.
(especiall
y hen
varying
over alarge
scale)
ratio
scale
log"
normal,gamm
a,eponential,
etc. (usually
a skeeddistr
ibution)
multip
licativ
e
!ll statistics
permitted for
interval scales
plus the
folloing$geo
metric
mean,harmoni
c
mean,coefficie
nt of variation
generali+ed
linear
modelithlo
garithmiclin
k
;:a)ETHICS IN RESEARCH
5esearch that involves human subjects or participants raises unique and comple ethical,
legal, social and political issues. 5esearch ethics is specifically interested in the analysis ofethical issues that are raised hen people are involved as participants in research. There arethree objectives in research ethics. The first and broadest objective is to protect human
participants. The second objective is to ensure that research is conducted in a ay that
serves interests of individuals, groups andAor society as a hole. 3inally, the third objective
is to eamine specific research activities and projects for their ethical soundness, looking atissues such as the management of risk, protection of confidentiality and the process of
informed consent.
3or the most part, research ethics has traditionally focused on issues in biomedical research.The application of research ethics to eamine and evaluate biomedical research has been
ell developed over the last century and has influenced much of the eisting statutes and
guidelines for the ethical conduct of research.
7/26/2019 BR Solution Set 2013
23/25
5esearch ethicists everyhere today are challenged by issues that reflect global concerns in
other domains, such as the conduct of research in developing countries, the limits of
research involving genetic material and the protection of privacy in light of advances intechnology and Internet capabilities.
;:b) ! proper research report includes the folloing sections, submitted in the order listed,
each section to start on a ne page. Some journals request a summary to be placed at the end of
the discussion. Some techniques articles include an appendi ith equations, formulas,
calculations, etc. Some journals deviate from the format, such as by combining results and
discussion, or combining everything but the title, abstract, and literature as is done in the
journal Science. 5eports ill adhere to the folloing standard format$
Title
!bstract
Introduction
8perimental etails or Theoretical !nalysis
5esults
iscussion
'onclusions and Summary
5eferences
Tit%e +nd Tit%e P+9e
The title should reflect the content and emphasis of the project described in the report. It should
be as short as possible and include essential key ords.
A$/t"+t
The abstract should, in the briefest terms possible, describe the topic, the scope, the principal
findings, and the conclusions. It should be ritten last to reflect accurately the content of the
report. The lengths of abstracts vary, but seldom eceed 6**"-** ords.! primary objective of
an abstract is to communicate to the reader the essence of the paper.
Int"#dti#n
E! good introduction is a clear statement of the problem or project and hy you are studying it.E
The nature of the problem and hy it is of interest should be conveyed in the opening
paragraphs. This section should describe clearly but briefly the background information on the
7/26/2019 BR Solution Set 2013
24/25
problem, hat has been done before (ith proper literature citations), and the objectives of the
current project. ! clear relationship beteen the current project and the scope and limitations of
earlier ork should be made so that the reasons for the project and the approach used ill be
understood.
E:pe"i&ent+% .et+i%/ #" T!e#"eti+% An+%/i/
This section should describe hat as actually done. It is a succinct eposition of the laboratory
notebook, describing procedures, techniques, instrumentation, special precautions, and so on. It
should be sufficiently detailed that other eperienced researchers ould be able to repeat the
ork and obtain comparable results.
In theoretical reports, this section ould include sufficient theoretical or mathematical analysis to
enable derivations and numerical results to be checked. 'omputer programs from the public
domain should be cited. &e computer programs should be described in outline form.
If the eperimental section is lengthy and detailed, as in synthetic ork, it can be placed at the
end of the report or as an appendi so that it does not interrupt the conceptual flo of the report.
Its placement ill depend on the nature of the project and the discretion of the riter.
Re/%t/
In this section, relevant data, observations, and findings are summari+ed. Tabulation of data,
equations, charts, and figures can be used effectively to present results clearly and concisely.
Schemes to sho reaction sequences may be used here or elsehere in the report.
.i///i#n
The cru of the report is the analysis and interpretation of the results. >hat do the results meanN
7/26/2019 BR Solution Set 2013
25/25
Citin9 Refe"ene/
%iterature references should be collated at the end of the report and cited in one of the formats
described in The !'S Style =uide or standard journals. o not mi formats. !ll references
should be checked against the original literature. &ever cite a reference that you have not read
yourself. ouble check all journal year, volume, issue, and inclusive page numbers to insure the
accuracy of your citation.