This image lists the various stages of the scientific method.
Quantitative Methods
Quantitative methods are generally useful when a researcher seeks to study large-scale patterns of
behavior, while qualitative methods are often more effective when dealing with interactions and
relationships in detail . Quantitative methods of sociological research approach social phenomena
from the perspective that they can be measured and quantified. For instance, socio-economic
status (often referred to by sociologists as SES) can be divided into different groups such as
working-class, middle-class, and wealthy, and can be measured using any of a number of
variables, such as income and educational attainment.
Qualitative versus Quantitative Methods
These two researchers are debating the relative merits of using qualitative or quantitative methods
to study social phenomena such as the learning processes of children.
Qualitative Methods
Qualitative methods are often used to develop a deeper understanding of a particular
phenomenon. They also often deliberately give up on quantity, which is necessary for statistical
analysis, in order to reach a greater depth in analysis of the phenomenon being studied. While
quantitative methods involve experiments, surveys, secondary data analysis, and statistical analysis,
qualitatively oriented sociologists tend to employ different methods of data collection and
hypothesis testing, including participant observation, interviews, focus groups, content analysis,
and historical comparison.
Qualitative sociological research is often associated with an interpretive framework, which is more
descriptive or narrative in its findings. In contrast to the scientific method, which follows the
hypothesis-testing model in order to find generalizable results, the interpretive framework seeks to
understand social worlds from the point of view of participants.
Although sociologists often specialize in one approach, many sociologists use a complementary
combination of design types and research methods in their research. Even in the same study a
researcher may employ multiple methods.
Identify data sources
Sources of information vary and can be classified as either internal or external Internal includes:
Organisational files, reports and/or surveys
Work intranet and other organisational material
Other staff
External includes
Government departments and publications
Representative associations such as employer bodies, institutes, unions and professional
associations
Newspapers and magazines (especially industry specific)
Television, radio and/or internet
Local and specialist libraries
Common methods of collecting information include
Interviewing colleagues/customers
Obtaining information from other organisations
Accessing previous file records
Observing and listening
Checking written material including referrals and client files
Questioning (in person or via indirect means)
Undertaking internet research
Checking research provided by others (in journals)
Validity and reliability
Information sources need to be evaluated for their validity and reliability. Validity the ability of a
scale to measure what it is supposed to. Reliability the degree to which measures are free from
error and yield consistent results. Note that Validity = Accuracy and Reliability = Consistency
Example Wikipedia. Is it a valid source? Is it a reliable source? Potentially no. Why? Wikipedia is a
community content website. Any member of the public can contribute content to the website so
there is a chance that the content is not valid or reliable.
When assessing validity and reliability consider:
The nature of the organisation undertaking the research
The purpose of the research and the sponsors of the research
The number and types of respondents involved in the research
The research instrument used to collect the information
There are various types of samples. A probability sampling is one in which every unit in the
population has a chance (greater than zero) of being selected in the sample, and this probability
can be accurately determined. Nonprobability sampling is any sampling method where some
elements of the population have no chance of selection or where the probability of selection can’t
be accurately determined. Examples of types of samples include simple random samples, stratified
samples, cluster samples, and convenience samples.
Good data collection involves following the defined sampling process, keeping the data in time
order, noting comments and other contextual events, and recording non-responses. Errors and
biases can result in the data. Sampling errors and biases, such as selection bias and random
sampling error, are induced by the sample design. Non-sampling errors are other errors which can
impact the results, caused by problems in data collection, processing, or sample design.
There are two sources of information from where data for marketing research can be
obtained17.
(A) Internal Sources:
These refer to the sources of information within the organisation. In certain cases internal
sources are indispensable without which the researcher cannot obtain desired results. Internal
sources include accounting information (Trading Profit & Loss A/c and Balance Sheets of
different years), salesmen’s reports, statistics in relation to advertisement expenditure,
transportation costs etc. Information from internal sources is easily available and no financial
burden is involved in gathering the information.
(B) External Sources:
In order to study marketing problems in detail the need of external sources of marketing
research arises. External sources are of immense importance and utility in case where research
needs detailed and thorough investigation. External sources data can be divided with two
categories (a) Primary data (b) Secondary data.
Primary Data:
This refers to the information collected by the researcher from original sources. It is not a
published data; it has to be gathered by the researcher himself by tapping various resources.
Primary data is usually collected for specific purposes.
The main sources from where primary data can be obtained are (a) Salesmen (b) Dealers; (c)
Consumers etc. It is a very slow process of collecting data and involves huge costs. But results
obtained from this data are original and tend to be more accurate and reliable.
These sources are explained below:
(a) Salesmen:
17 Source: Your Article Library, as at http://www.yourarticlelibrary.com/marketing/sources-ofmarketing-
research-a-internal-b-external-sources-of-marketing-research/25860/, as on 4th September,
2017.
Salesmen are the most important source of providing first hand information. They are
appointed by the owners for the sale and promotion of its products. They have a direct link with
the consumers understand tastes, preferences and buying habits of the consumers.
They can also know about the dealer’s reaction (especially of retailers) towards the firm’s
products by taking into consideration price, design, packaging and size etc. of the product. The
marketing manager may direct the salesmen to prepare periodical reports containing the
information collected by them.
The information collected in this manner is original and more meaningful. This will further
enhance the morale of salesmen as they feel that they are contributing towards the formulation
of marketing policies of the organisation.
But sometimes information provided by salesmen is not accurate and upto the mark. The
salesmen are not properly trained and do not know the methodology to collect the information
properly.
(b) Dealers:
This is another source of collecting primary data. Valuable information can be collected with
regard to demand of the product from retailers. Information about the marketing policies of
competitors can also be gathered from the dealers.
It has been observed that sometimes this method does not prove to be fruitful as dealers do
not keep proper records and they do not want to waste their time in supplying information.
(c) Consumers:
This source of collecting primary data is of great importance. Representative samples of
consumers may be selected for conducting thorough investigation with regard to price, quality
and use of the product. This method of collection data is very reliable as it establishes direct link
between producer and the consumer.
Secondary Data:
Secondary data is already existing which has been collected and published by some individuals
or institutions. This data is available at a very low cost and it requires lesser time to collect it.
The main sources of secondary data are:
(a) Periodicals and Newspapers:
Business magazines and journals published periodically contain data which is very useful for
marketing research; Newspapers such as Economic Times and Financial Express also contain
data regarding business trends and market reports. Important trade journals are Industrial
Times, Commerce, Capital, Market, Indian Finance, Business India, Business World and others.
(b) Govt. Publications and Reports:
There are innumerable publications brought by Central and State Govts, which contain valuable
data for conducting marketing research. Census reports of the Government of India,
Publications of Planning Commission; periodical publication such as Indian Review, various
markets bulletins.
Reserve Bank of India bulletin, publication of the Statistical Departments of various State Govts.,
supply valuable information extensively used in marketing research.
(c) Trade Associations:
Various trade associations like Chambers of Commerce, Export Promotion Council etc, publish
useful data which is of immense help to the res warmer.
(d) Published Surveys Of Markets:
This is another useful source of supplying secondary data. Market surveys and reports are
important instruments in the hands of researcher for conducting marketing research. These are
published by business houses or independent research organisations. These pertain to specific
lines of products.
(e) Foreign Govts, And International Agencies:
Publications of foreign Govts, with regard to trade and other important aspects of economy of
respective countries and information published by UNO, ILO, IBRD (International Bank for
Reconstruction and Development) serve useful purpose in making comparison of Indian
conditions prevailing in other countries of the world.
(f) Other Sources:
Besides the above mentioned sources of marketing research, there are many other sources of
supplying secondary data e.g., colleges and universities stock exchanges and commodity
exchanges, specialised libraries’, internal sources such as sales and purchase records, salesman,
reports, sales orders, customer complaints and records of other companies.
The secondary data collected from above mentioned sources suffer from certain limitations.
The basis undertaken by different agencies for collecting data may not be comparable. In other
words, uniform basis may not be adopted for data collection. The data may be based on
incomplete records under secondary source; data is collected for purposes other than
marketing research.
Quantify required data
Common Types of Market Research18
By “types of research” I don’t necessarily mean how you’re collecting data. Instead I mean the
procedures and methodology used to analyze the data collected.
These procedures include market segmentation, product testing, advertising testing, key driver
analysis for satisfaction and loyalty, usability testing, awareness and usage research, and pricing
research (using techniques such as conjoint analysis), among others.
Market Segmentation
When conducting market segmentation studies we’re generally asking survey questions aimed at
capturing needs, values, attitudes, behaviors and demographics. A B2B company might also want
to investigate firmographic data such as company size, revenues, and product category that are
relevant to the industry in question.
Marketing can’t effectively speak to every type of person or business at the same time, so one of
the main goals of market segmentation is to allow for more efficient and effective marketing
tactics.
Without market segmentation companies are shooting in the dark and wasting valuable bullets.
They may hit a customer by accident, but they would miss a lot of others.
Product Testing
A detailed understanding of how your product meets (or doesn’t meet) your customer’s needs is
crucial both to product development and marketing, so these types of market research studies
need to be conducted throughout a product’s life.
Ultimately you should be able to make informed “go” or “not go” decisions about new features
and products before launch, and thus save capital, time and effort.
Successful product testing should:
Give insight into product/service viability by investigating competing and substitute
alternatives along side customers’ willingness to embrace new products/services.
18 Source: Survey Gizmo, as at https://www.surveygizmo.com/survey-blog/market-research-types/,
as on 4th September, 2017.
Determine competitive advantage as well as possible threats from similar
products/services.
Identify the products with the highest revenue potential.
Clarify what improvements should be prioritized before a product launch (or re-launch).
Pinpoint which product features (both existing and potential) are most important to your
target audience.
Help produce marketing messages to change or enhance existing perceptions about your
products/services.
Advertising Testing
Like product testing, tests of your advertising campaigns can save you valuable time and
resources. By taking potential campaigns directly to your audience and gauging their response you
can focus on creating truly impactful advertising.
Satisfaction and Loyalty Analysis
Satisfied customers aren’t necessarily loyal customers, but consistently measuring customer
satisfaction is a great way to increase customer retention.
This type of research is aimed at identifying key drivers of satisfaction and measuring the likelihood
of customers to continue using a company’s products and services.
The goals of these kinds of studies are:
Determine what factors influence loyalty, advocacy, and repeat purchases, including
product/service attributes, company operation, customer service, price, etc.
Carefully monitor overall satisfaction, recommendation likelihood, and defection likelihood
over time.
Provide early warnings about emerging gaps in product/service performance, customer
service, and processes that might lead to customer defection.
Help identify areas of the product or service that need improvement to meet changing
needs.
Guide the creation and/or ongoing development of customer loyalty and retention
programs.
Signal when organizational changes need to be made to improve operations and
customer retention.
Brand Awareness and Reach
By conducting regular, well-designed brand awareness surveys you can keep tabs on how effective
your marketing campaigns really are.
When done right, a brand awareness survey can help you measure:
Brand Recall: Can a customer spontaneously recall your brand, or do they think first of a
competitor?
Brand Recognition: When presented with a list of brands, does your audience recognize
yours as a reputable option?
Brand Identity: Brand identity is what you as a marketing team create. It’s important to
determine whether these efforts are being successful.
Brand Image: While brand identity is created by the brand itself, a brand’s image is based
in the customer’s perception alone. Tracking disparities in these two can reveal gaps in
your marketing efforts.
Brand Trust: In an era of data breaches, keeping tabs on your levels of brand trust is key. If
your brand doesn’t appear trustworthy, you will have difficulty retaining customers.
Brand Loyalty: Loyal customers can become evangelists, but you need to consistently track
loyalty levels to determine how often this transformation is happening.
Customer Profile: Changes in your core customer base may signal the need for a pivot,
either in the product or your marketing messages (or both).
.
Pricing Research
Surveys that ask customers to choose between different products with unique features and price
points, typically done via conjoint analysis, can help you identify what features are most valuable to
your audience and what they’d be willing to pay for them.
Combined with some basic research on your competitors’ pricing, these insights can give you a
distinct advantage in pricing your products and services.
Knowing Which Type of Research to Use
When to use each of these different types of market research data collection methods and types of
research depends on the business issues we are dealing with in one or more of four key areas:
Awareness: let the market know that the product or service exists
Targeting: reach the target segments with the highest profit potential
Acquisition: optimize the marketing message, offer, and price that will close the sale
Retention: generate repeat purchases from current customers
The chart below, which we call the Relevant Wheel, shows when it is most appropriate and relevant
to conduct different types of research.
Data Collection and Analysis Methods
Here we need to make a distinction between data collection methods and market research types
based on analytical approach, which are often confused. Data collection methods differ based on
whether we want to conduct quantitative or qualitative research.
Qualitative research, which is exploratory in nature, usually uses data collection methods such as
focus groups, triads, dyads, in-depth interviews, uninterrupted observation, bulletin boards, and
ethnographic participation/observation.
Quantitative research, which looks to quantify a problem, collects data through surveys in different
modalities (online, phone, paper), audits, points of purchase (purchase transactions), and clickstreams.
Choose the Market Research Type that Meets Your Needs
Next time you wonder what type of market research to conduct, I invite you to ask yourself where
the particular problem at hand belongs: Awareness, Targeting, Acquisition or Retention. Then take
a look at the Relevance Wheel to find the approach that will help you answer your specific
questions.
If you choose your method carefully market research can give you a big advantage over your
competition.
Analysis of your research in terms of what statistical information is presented is a common and
powerful exercise. Statistical analysis allows the researcher to view data to determine trends,
frequency, relationships and commonality. Most statistical methods used in business are known as
inferential statistics. That is, using a sample to infer the results across the rest of the population. If
one group of people behave/think this way, then it’s likely that everyone else will. Commonly used
statistical analysis measures:
Mean, median and mode
Variance and standard deviation
Probability Testing hypotheses
Correlation
Regression analysis (to predict future results) Mean median and mode
The mean (or average) of a set of data values is the sum of all of the data values divided by the
number of data values. The median of a set of data values is the middle value of the data set when
it has been arranged in ascending order. That is, from the smallest value to the highest value. The
mode of a set of data values is the value(s) that occurs most often. Variance and standard
deviation: are measures of how spread out a distribution is. They measure how variable a group of
results are. Standard deviation is a measure of how spread out your data is. Computation of the
standard deviation is a bit of a process.
Compute the mean for the data set
Compute the deviation by subtracting the mean from each value
Square each individual deviation
Add up the squared deviations
Divide by one less than the sample size
Take the square root
Example You are being asked to look at a local coffee shop. The owner wants to make some
decisions about the business and doesn’t know where to start. Among other things you decide to
survey 10 people about the coffee and take a score out of five. You get your results and find that
the mean is three.
You do however need to understand the spread of these figures. The results mean something
quite different given that there are a few fives and ones. Finding the standard deviation shows us
how spread out the numbers are. This helps us determine what is really going on. If you only relied
on the mean, three, you would think things were fine, slightly above average. Ones however
suggest problems, knowing the spread is very useful. You can obviously see the spread on such a
small sample, but imagine you surveyed 10 000 coffee drinkers.
Variance = the average squared deviation of each number from its mean
Probability – the measure of how likely it is for something to happen.
Testing hypotheses – A hypothesis is an assumption or proposition that a researcher makes about
some characteristic of the population being investigated?
What is the problem?
What is your expectation?
Create a model to measure the data against
Use your model to guide data collection
Compare your expectation with the information at hand (data)
Use the data to modify your model
Use the information and model to make predictions about the hypothesis
Correlation: the degree and type of relationship between two variables in which they vary together
over a period of time.
For example, the Grade Point Average (GPA) achieved at school in relation to the hours of
television watched.
Regression analysis
Predicts the outcome of an event (the dependent variable) based on interactions between other
related drivers (the explanatory variables). It is important to note that it is difficult to be sure as to
what are explanatory variables (or causal), and what happen to be coincidence. Causality is difficult
to establish. Just because A increases at the same time as B decreases, does not prove a
relationship between the two. For example, you can reasonably predict sales volumes by looking at
marketing expenditure and the number of sales representatives your company employs. As
marketing expenditure and sales rep numbers increase, you can reasonably expect a
corresponding increase in sales volume.
What performance hypotheses did you come up with?
Does this mean that salary, gender or tenure cause performance? No, not necessarily. There are
many possible explanations for the suggestions of relationships. For example:
Barbara gets paid more than anyone else in the group. She may be the Sales Director, and
therefore have more responsibilities as well as a smaller sales region than the rest of the
team
Janet, while earning the least amount and has the second lowest performance %, may only
work part time
Henry may have the region with the largest client base, and therefore does not need to
work particularly hard to make sales
You will need to make assumptions when analysing the data you gather. Part of the research
process is ‘sanity checking’ your assumptions with others. Peer and expert review is critical to your
success.
When checking your assumptions validity, ask yourself…
Can I demonstrate a clear relationship between the data and my conclusions?
Is there any other explanation for this relationship?
Can I explain how I came to my conclusion?
Do my assumptions relate to my research objectives?
Quantifying Customer Value19
A Value Proposition answers 3 central questions:
Is our offering better?
In what ways is it better and how much better is it?
What is that worth to the customer?
To be useful in sales conversations, a good Value Proposition boils the answers to these questions
down to the two to four ways that our offering is better than the competition and what that is
worth to the customer. A good Value Proposition quantifies these differentiators in a way that is
clear and simple, cutting through technical complexity.
19 Source: Leverage Point, as at https://www.leveragepoint.com/blog/quantify-customervalue/
quantifying-customer-value-good-data/, as on 4th September, 2017.
A good framework helps provide clear thinking in quantifying customer value:
What are we offering?
To what customer segment?
Compared to what reference point?
Our offering’s positive differentiation creates value for that customer that can be highlighted and
articulated as value drivers.
It doesn’t take a rocket scientist or an MBA to set up a value driver. Common sense and a desire to
understand the customer’s business are usually enough. The quantified value of a differentiated
product or offering is built using the following elements:
A formula
Data to quantify our product’s claimed benefit
Data about market conditions and the customer’s operations
Let’s look at each of these elements in turn to understand how to start and how agile marketing
results in ongoing improvement.
Value Driver Formulae. A good value driver formula is based on an understanding of the
customer’s business, turned into a formula through straightforward arithmetic. Good product
managers start with simple questions:
How does my offering reduce my customer’s costs?
How does my offering increase my customer’s revenue and profitability?
How does my offering reduce customer risk and what is that worth?
The value driver formula picks out key elements of the customer’s market and operations as a
basis for quantifying value, using data that capture product claims to provide a clear comparison
between a competitive reference point and our offering. These formulae are invariably created by
teams who generate hypotheses which they test with experts, experienced sales reps and/or
selected customers. Formulae usually start simple. They often evolve to a more detailed and
analytical form as the team gets deeper answers from their tests, understanding the customer
better. Simplifying and suppressing detail is necessary to prepare the Value Proposition for sales
use and effective communication.
The formula itself is a way to present important information. It can be displayed in a way that
highlights key data and shows clear calculations transparently. Or it can be complex, obscure, hard
to follow, shown with too much information or include irrelevant detail. An agile marketing process
should consider how formulae are presented, test alternate approaches and improve the
presentation of the Value Proposition based on sales experience. Suppressing detail can help
improve the clarity of most conversations. Structuring formulae and initial formula display simply to
support these conversations is usually good.
The post Scientific Method is Essential Tool in Research appeared first on My Assignment Online.