The researcher’s next task is to make sense of the collected data. Before the researcher can gain
understanding from the collected data, he/she must first examine the raw information (i.e., what
was actually collected) to make sure the information exists as required. There are many reasons
why data may not be presented in the form needed for further analysis. Some of reasons include21:
Incomplete Responses – This most likely occurs when the method of data collection (e.g.,
survey) is not fully completed, such as when the person taking part in the research fails to
provide all information (e.g., skips questions).
Data Entry Error – This exists when the information is not recorded properly which can
occur due to the wrong entry being made (e.g., entry should be choice “B” but is entered
as choice “C”) or failure of data entry technology (e.g., online connection is disrupted
before full completion of survey).
Questionable Entry – This occurs when there are apparent inconsistencies in responses
such as when a respondent does not appear to be answering honestly.
21 Source: Know This, as at https://www.knowthis.com/planning-for-marketing-research/step-5-
evaluate-data, as on 4th September, 2017.
To address these issues the researcher will take steps to “cleanse” the data which may include
dropping problematic data either in part (e.g., exclude a single question) or in full (e.g., drop an
entire survey). Alternatively, the research may be able to salvage some problem data with certain
coding methods, though a discussion of these is beyond the scope of this tutorial.
After evaluating questions and information requirements, the next step is to consider what data
sources and methods are likely to yield the necessary information and to develop a research plan.
This requires:
Which existing data sources are suitable
What gaps in information are there and are they critical
Method of information collection
Achievability of the research plan given available resources and timing of the evaluation
Privacy and ethics approval requirements
Effective data sampling
Data sampling is simply the process of using a small percentage of a total population to assist you
to understand how the entire population might respond. Often, conducting a comprehensive
survey of all staff of a company, all residents of a suburb, all citizens of a country or even all
members of a team is difficult, costly, time consuming or simply impossible.
Selecting a sample procedure
After determining whether using a sample or a population would be appropriate to use for our
given survey, the next step is to determine what sampling procedure to use. Selecting a sampling
procedure consists of two major groups:
Probability sampling
Non-probability sampling
Probability sampling consists of the following procedures:
Simple random sampling – each member of the population has a known and equal chance of
being selected e.g. Lotto, drawing lots.
Systematic sampling – each member has a known but not equal chance e.g. randomly select a
number (1 to 10), if the number 5 is picked, then every fifth number in the list is selected. (Bias)
Stratified sampling – the population is divided into strata (group levels). Each stratum has its
common and measurable characteristics (age, income, geographical area etc.), and a random
sample is done for each stratum. (Time-consuming)
Non-probability sampling consists of the following procedures:
Judgment sampling – occurs when a knowledgeable person selects sample members that they
believe are appropriate. (Subjective) Convenience sampling –based on the researcher’s
convenience, i.e. location.
Quota sampling – is similar to the stratified sampling method, whereby, an appropriate number in
each stratum is selected e.g. universities use this system for some of their courses, such as
physiotherapy.
Snowball sampling – is a form of judgmental sampling that is very appropriate when it is necessary
to reach small, specialised populations. Initially a group is selected (via the judgment sampling
method) specifically because they purchased the product.
To convert your data into a manageable form, sort and summarise so it is easier to analyse. Ways
to do this include drawing attention to key pieces of information by:
Adding notes or comments to data
Highlighting sections of text or figures
Labelling material
Selecting particular sections of data and filing them together
Choosing a few typical examples to describe your larger body of data
Noting critical or important pieces of information
Data processing and analysis
Data analysis is the process where raw data is organised so useful information can be extracted
from it. The process of organising and thinking about data are critical to understanding what the
data does and does not contain. There are a variety of ways in which people can approach data
analysis, and it is very easy to manipulate data during the analysis phase to push certain
conclusions or agendas. It is important to pay attention when data analysis is presented, and to
think critically about the data and conclusions which were drawn. Remember to always be mindful
of:
Ultimately, poor information will likely result in a poor decision
Raw data can take a variety of forms, including measurements, survey responses and observations.
In its raw form, this information can be incredibly useful, but also overwhelming. Over the course
of the data analysis process, raw data is ordered in a way which will be useful.
For example, survey results may be tallied, so that people can see at a glance how many people
answered the survey, and how people responded to specific questions. In the course of organising
the data, trends often emerge, and these trends can be highlighted in the research report so that
readers take note. In a casual survey of ice cream preferences, for example, more women than
men might express a fondness for chocolate, and this could be a point of interest for the
researcher. Modelling data with the use of mathematics and other tools can sometimes exaggerate
such points of interest in the data, making them easier for the researcher to see.
Methods of analysis
Data sampling is the process of using a small portion of the population to help you understand
how the whole population may react. Examples include political polls, television ratings data and
what housewives think of a new washing powder. Feedback on results of research you have started
from interested parties is invaluable to make sure your research is relevant and useful. Interested
parties can include managers, decision makers, experts and project owners. Peer review –
Providing results or preliminary findings to your peer group can be as useful as getting feedback
from interested parties. Peers can provide you with different ideas and insights into your research,
and can suggest additional sources of information.
Review of previous research others have done before you is critical to the success of nearly all
research projects. Whether you use other’s research as a primary source or just to develop
research strategies, it is essential to be careful in choosing past research.
With the data in a form that is now useful, the researcher can begin the process of analyzing
the data to determine what has been learned. The method used to analyze data depends on
the approach used to collect the information (secondary research, primary quantitative research
or primary qualitative research). For primary research the selection of method of analysis also
depends on the type of research instrument used to collect the information.
Essentially there are two types of methods of analysis – descriptive and inferential.
Descriptive Data Analysis
Not to be confused with descriptive research, descriptive analysis, as the name implies, is used
to describe the results obtained. In most cases the results are merely used to provide a
summary of what has been gathered (e.g., how many liked or dislike a product) without making
a statement of whether the results hold up to statistical evaluation. For quantitative data
collection the most common methods used for this basic level of analysis are visual
representations, such as charts and tables, and measures of central tendency including
averages (i.e., mean value). For qualitative data collection, where analysis may consist of the
researcher’s own interpretation of what was learned, the information may be coded or
summarized into grouping categories.
Inferential Data Analysis
While descriptive data analysis can present a picture of the results, to really be useful the results
of research should allow the researcher to accomplish other goals such as:
Using information obtained from a small group (i.e., sample of customers) to make
judgments about a larger group (i.e., all customers)
Comparing groups to see if there is a difference in how they respond to an issue
Forecasting what may happen based on collected information
To move beyond simply describing results requires the use of inferential data analysis where
advanced statistical techniques are used to make judgments (i.e., inferences) about some issue
(e.g., is one type of customer different from another type of customer). Using inferential data
analysis requires a well-structured research plan that follows the scientific method. Also, most
(but not all) inferential data analysis techniques require the use of quantitative data collection.
As an example of the use of inferential data analysis, a marketer may wish to know if North
American, European and Asian customers differ in how they rate certain issues. The marketer
uses a survey that includes a number of questions asking customers from all three regions to
rate issues on a scale of 1 to 5. If a survey is constructed properly the marketer can compare
each group using statistical software that tests whether differences exists. This analysis offers
much more insight than simply showing how many customers from each region responded to
each question.
Make decisions on data types, combinations, gathering methods, sources,
quantities and processing methods
Choices achievable with available resources
Cost benefit of choices
Are the choices consistent with organisational procedures, policies?
Will the choices satisfy the research objective?
Developing conclusions
In order for your research to be useful, you need to be able to draw some conclusions. You need
to:
Make sense of your research material
Form an opinion about the data gathered
Develop a line of reasoning
Make recommendations as to what should happen now
You should be able to:
Write your conclusions clearly and succinctly
State how they relate to the research and business objectives
Clearly point to the evidence that your conclusions are logical and based on the available
evidence
State your conclusions objectively and without bias
Data processing options
Your business has a range of options to manage the data processing:
Data processing service experts
Manual or personal methods
Packaged analysis routines or programs
Specialist software packages
It is easy to fall into traps when analysing or processing data yourself, particularly if you have an
interest in the outcome. It is better for the reliability of the analysis to engage an unbiased third
party or to use specialist programs. However, this can be quite costly. When deciding which option
is right for your research project, it is important to consider:
The budget
Timeframes
Expertise available
Technology available
Purpose of the research
The impact on business and the critical nature of the research
A simple way to analyse the above is a grading system answering key questions:
How easy will it be to implement the research project?
Is the methodology of a sufficiently high quality and credibility?
How useful is the research likely to be?
Are alternative information results available with the same or higher level of validity and
reliability and with lower resource costs?
Examples of Marketing Research Problems
Companies and other organizations use marketing research to manage the risks associated with
offering new products and services. These organizations don’t want to spend too much money
developing a product line that research indicates will be unsuccessful. Some problems make
marketing research costly and produce results of questionable value for the organization.
Survey Design
Organizations use marketing research to find out what customers think and what they want.
The survey is a direct way of collecting quantitative, or numerical, information and qualitative, or
descriptive, information. When there are errors in the survey design, marketing research
problems can surface. For example, a company might use a method that is designed to collect
a random sample from the target consumer population, but the method is not really random.
Therefore, the organization cannot generalize its survey results to represent the target
population.
Survey Nonresponse
One marketing research problem relates to how the survey is offered to the target population.
Marketers design a survey that many customers choose not to respond to. They look at reasons
why people don’t want to participate, and they might reach conclusions such as the survey
takes too much effort or that the incentive for participation is not appealing to respondents.
Survey Bias
A survey might include one or more sources of bias. Marketers might believe, for example, that
they have created an online survey to appeal to respondents of all ethnic backgrounds, but the
survey questions, and even images, might be biased to favor one ethnic group or could offend
one or more ethnic groups. A survey’s format and content must be acceptable to all audiences
from which marketers seek to gather information.
Observation Research
Some marketing research involves observing consumers in action and noting their preferences.
Marketers can become intrusive, interfering with a consumer’s experience to the point that a
consumer feels disgusted and leaves the business site. For example, a fast-food chain’s
researchers need to determine if there is a need for a new location of its store so they survey
people going through the drive-through line. Although researchers conduct a short survey,
they aggravate customers by slowing down the line.
The post Identify and evaluate suitable data processing methods appeared first on My Assignment Online.
