C5

Question 1

Why do secondary research?

Answer 1

Almost all research projects depend upon or can benefit from a secondary research stage. This is especially useful in early stages of project, in order to clarify thinking, help with problem definition, and with research design and planning. It can also act as context or framework for analysis or interpretation of findings.
It is possible that 2ndry research may entire set of research objectives alone with no need for primary research.
Value or usefulness of any piece of research or set of data is rarely exhausted on initial or primary application and may be useful in same context at a later date or in a different context. One set of data e.g. geodemographic database, may be combined with others making the combined set more valuable and of greater use than the individual elements.

Question 2

List uses and benefits of 2ndry rsearch?

Answer 2

Answers research problem without need for primary research
Better understanding of issues and wide context of problem
Helps define problem
Helps in development of hypothesis
Determines nature of evidence required to address the problem
Helps to formulate effect research design
Helps in define of effective questions
Enhances analysis and interpretation of 2ndry data
Sets findings from primary research into a wider context
Existing data may be cheaper
Secondary data are relatively quick and easy to get a hold of

Question 3

General process of secondary research?

Answer 3

Identifying existing relevant sources of data / information
Gaining access to them
Assessing their suitability for purpose
Reviewing them and assessing their quality
Using knowledge created

Question 4

Why use a literature review?

Answer 4

Literature reviews are commonly uses in early stages of project or at analysis and interpretation stage. This is valuable as it can be used to increase knowledge and understanding of topic under investigation; allows you to benefit and build upon the work and thinking of others e.g. those who have tried a similar research approach or those who have done research on the same or similar topic and so should enhance quality of thinking and research. Cam be referred to as a synopsis or critical assessment of relevant literature.

Question 5

what should loterature review show?

Answer 5

What is known about topic
What you have to say about what is known
If anyone else has conducted the same or similar research
If anyone has done research that is related to your topic
Where your research fits in with what exists already
Why research is worth doing in light of what has been done already

Question 6

process of literature review?

Answer 6

Define purpose of review precisely to avoid being overwhelmed by large amount of irrelevant material
Determine what sources will be useful e.g. books, government publications, commercial research reports, journal articles, conference papers, blogs. This will dictate where you should be search for material
Prepare list of key words, relevant dates to form basis of search in search engines and online catalogues
Set up database or spreadsheet to record sources and information you find; you may want to devise a template in which you can make notes on articles and folders to store relevant, useful material or material you are unsure about
Assess whether material is relevant and useful and therefore should be read in-depth and critically - scan / read through abstract or if they do not have one at aims and objectives, methods and conclusions. Get rid of irrelevant material
Review and critically evaluate each piece. Look at strengths, weaknesses, robustness of research design, relation to your project, how information from material can add value to your project

Question 7

Where do you find secondary sources of data? general

Answer 7

In marketing context 2ndry resources are often classified according to whether they are internal to organisation, produced by it, or external. Internal sources include reports / data from previous research, financial data, sales data and customer data. External sources are those produced by others outside of the organisation and can include reports / data produced by government bodies, market reports by commercial organisations such as Mintel, geodemographic databases compiled by companies such as CACI and Experian, articles published in journals such as International Journal of Market Research, online discussions or blogs by people within particular market or sector.

Question 8

How does EPOS work?

Answer 8

Data can be captured at point of interaction with customer, whether in person or EPOS - electronic point of sale, data collected scanners or remotely by the internet - stored in and retrieved from databases and data warehouses designed to function as management information systems (MIS) or marketing information systems (MkIS) - such systems are often referred to as decision support systems (DSS) or executive information systems (EIS) and are structured in a way that allows users to search for and retrieve data needed for planning, control and strategy development

Question 9

How do loyalty / rewards cards work? Pros and cons?

Answer 9

Loyalty / rewards cards - link personal data with buying behaviour at level of individual customer. Customers apply for cards to benefit from organisation promotion schemes, when doing so they enter personal details, each time a card is used customers personal details are logged against transaction / purchases made. Organisation then has a record of your actual buying behaviour when using cards. Same process is used online when you register to use a site - where you agree to having your details and activity recorded and stored
Cards however do not offer data on behaviour outside of purchases used with card, demographic data may not be accurate, people may hold more than one card, not use for all purchases e.g. only large purposes, and is likely to be used by a specific kind of customer and so does not offer a representative sample of customer base

Question 10

Types of online behaviour tracking?

Answer 10

Online behaviour is tracked in four ways - logging IP address; storing cookies in computer; registering and recording ‘web bugs’ you trigger in the sites you visit; and by accessing browsing history.

Question 11

Cookies?

Answer 11

Storing cookies - bits of data - in your computer means that the website has a unique identifier for you so that when you visit again you are recognised. Cookies from a website may also allow you to be recognised by other related websites (monitored by the same company). Some cookies (third party used by advertisers) can monitor your activity across multiple websites and build a profile of browsing activity.

Question 12

IP address?

Answer 12

Your IP address is the address to which website content is sent, as soon as you go to a website it sees your IP address and can use this to find your geographic location. Unless IP address changes the website can monitor all activity from that address.

Question 13

Web bugs?

Answer 13

‘Web bugs’ are used to track who views website and from which IP address - they are a means of following you from website to another within a group or network of sites.

Question 14

Accessing browser history?

Answer 14

Websites can program code to look at your browsing history on your computer - once they find it they use it as the basis for categorising or profiling you, in order to personalise the websites you visit with relevant ads.

Question 15

What is purchase pattern analysis?

Answer 15

Purchasing patterns can be analysed e.g. shopping basket analysis, and on basis of past behaviour retailers can send recommendations and alerts on offers of the types of items or brands you buy. Retailer can also compare purchase records with other customers and use similarities to notify that ‘customers who bought this item also bought…’ SBA can also show sets of products or brands that are bought together which groups / segments their customers together allowing them to be targeted with tailored offers. Examining behaviour time can allow company to build models to predict behaviour, sales volumes and revenue. This information can be used to understand how profitable different groups of customers / outlets are, and which promotions and advertisements work for each group.

Question 16

What is analysis of databases reffered to? CCR

Answer 16

Analysis of databases to identify customer behaviour and sales patterns to build profiles of customers / outlets to identify market segments or gaps is classic confidential research where there is no feedback of personal data.

Question 17

External sources and databases?

Answer 17

external sources Can also be integrated into the organisation’s DSS (decision support system), and is data generated from other organisations. External sources are commonly sorted into two categories: official statistics produced by government departments, its agencies and related bodies and unofficial statistics produced by trade bodies, commercial research organisations and business publishers. There is data that does not fit into these categories though e.g. library books, scholarly journals, magazines, newspapers, statistical publications and full range of electronic resources.

Databases can be built from multiple sources / channels using for example demographic data and geographic data to build a geodemographic database or information system. This could be expanded to include financial data and lifestyle data.

Question 18

What are GIS?

Answer 18

Sometimes called geodems, are a form of 2ndry data often used in market research. They are used to build geodemographic classification / information systems which are then primarily used to identify and target different types of consumers.

Geodems relate demographic characteristics of residential population, derived from the Census, at the smallest geographical unit within the Census for which data is available, to geographic information about that area. Sources of information used to construct system and build categories may also derive from electoral register, postal address files, car registration information, credit rating data, data from surveys on media use or attitudes and data from customer databases. The end product is a classification of neighbourhoods or areas within which people certain characteristics live. The classification or segmentation will be based on cluster analysis. Each ‘cluster’ or type of neighbourhood will be different from every other cluster in neighbourhood because people living in that area will be different. A neighbourhood in one area may be classified to the same cluster or type of neighbourhood in another town, because characteristics are similar.

Question 19

What can GIS be used for?

Answer 19

Since they are rooted in geographic information, knowing a persons postcode is enough to allow you to assign them to a particular geodem group - allowing each customer on database to be assigned to gemdem cluster. Thus data from surveys on buying behaviour, attitudes and values or data from customer database can be linked. Databases created by this fusion allows examination of relationships between consumers, their attitudes and behaviours. Using geodem info can help organisations to gain an in-depth understanding of customers habits, preferences, attitudes and opinions. This can be used to develop strategy, target product, services and marcomms more effectively. Applying geodem information to sales data based on loyalty cards for example will help give demogs and lifestyle characteristics. You can then use GIS to find out where people with similar profiles are located and use this information to plan location of stores or business, type of store, size, products it should contain, targeting marketing and ad campaign

Question 20

Examples of GIS systems?

Answer 20

GIS systems are available for purchase, including CACI’s ACORN system and Experians MOSAIC.

Question 21

Sources of government data?

Answer 21

Govs and related bodies collect wide range of social, economic and business data, from Census of Population (demogs of population), through spending habits, lifestyle and attitudes, to information about different market sectors, agriculture and tourism to international trade and key economic indicators.

The Government Statistical Service, an organisation spread among most government departments and devolved administrations, collects, analyses and disseminates official statistics in order to meet needs of government, business and public’. Within the GSS is the role of the ONS, a body independent to any other government department, collates research and statistical publications produced by government departments and related bodies in compendia publications and databases.

Government statistical services exist in most countries, where government departments for trade and foreign affairs and embassies of foreign governments are useful sources of data on information on political, legal, economic and cultural aspects of business or research.

Question 22

What processes should you follow when citing sources or using material belonging to another person or organisation?

Answer 22

There are several ways of citing material - in Harvard System, you give authors surname and year of publication e.g. (Hakim, 1982), or if there is not an author you give name of organisation e.g. (MRS 2010). If writing formal literature review, you may need to follow the citation practices of the subject area, university or publication to which you are sending work. If you use a source you must acknowledge and cite said material, or this is plagiarism - which is a serious offence. When copying or downloading material you must abide by copyright laws. If you use materials from other sources in research, report it if it is your responsibility to get appropriate copyright permission to use them.

Question 23

What is consumer-generated content? How can it be used?

Answer 23

This is the content seen online blogs, newsgroups and social networking sites. It is material produced by other people in online conversations that contains information / opinions about organisations / products / services that marketers are interested in. as a result many organisations conduct social media monitoring or social media market research, gathering or ‘scraping’ content from relevant sites and analysing it. When aim is to find out what people are saying about your company or brand, say, a topic relevant to your organisation, it is referred to as buzz monitoring. Since it is observation and not interviewing, it is sometimes referred to as netnography or webnography.

Question 24

How does web-scraping work?

Answer 24

Sites are sampled by online directories or company’s own bespoke web scraping software, which is programmed to search sites for relevant key terms or words that consumers might use in relation to the organisation, brand or topic under investigation. Data is collected and structured into a readable and searchable format, stored in a database, cleaned and analysed to determine which elements of data are relevant. Data deemed relevant can be analysed by either a qual method or using quant techniques of data mining including pattern detection.

Question 25

What can web-scraping achieve?

Answer 25

seful for developing consumers view of a product, service or brand and the language they use to talk about it. It is useful in assessing image or reputation of brand, for tracking the impact of advertising and other marketing activity and for gathering intelligence about competitors.

Question 26

Explore ethics of using consumer-generated content

Answer 26

Researchers should respect internet users’ expectations of privacy, noting that researchers are not the intended audience. Researchers should not engage with commenters and should gain consent for sharing consumers comments in any research. Identities of commenters should be obscured. Typically, most data collected - mentions of organisations, brands, services and the surrounding text - would not fall into the category of personal data, and therefore conform to relevant terms of the MRS COC and DPA. If buzz monitoring, effort should be taken to avoid collecting personal data.

Question 27

What is the process of assessing quality and suitability of secondary research? CDV

Answer 27

Once a secondary source is found, it should be judged as to whether it is of sufficient quality to be used and whether it is useful for purposes of research. Quality and suitability go hand - you do not want to use research that is of bad quality, even if it is suitable and vice versa. In the first stages of your evaluation you should establish whether material has authority and credibility: ``` Find out who commissioned the research Who produced research Why it was commissioned Where it was found Do commissioners or providers belong to any research standard associations e.g. MRS ``` This is because the commissioner may have a particular agenda or approached research with a perspective that may bias research. The source of data may have a negative / positive reputation in research / data. Social media monitoring is not a sound source of secondary data, as you are not aware of context or why the content was posted or intended for. In second stage of assessment you should establish how up to date it is by: Finding out when work was done Assessing whether it relies on outdated facts or figures Whether it has been overtaken by recent discoveries or changes It is important to note that there may be a time lapse between when work or fieldwork was done and when it was published or uploaded to a data source. The last stage is to check validity, reliability and accuracy by: Assessing research design e.g. whether it was appropriate for producing valid and reliable evidence to answer research question Evaluate sampling procedure and whether it was appropriate to aims of research problem Looking at sample sizes and sub-samples to assess whether they are robust enough for any claims made Assess method of collection and whether it was appropriate Assess response rate and whether it is large enough to ensure a representative sample of population of interest Accuracy of data Studying quality standards employed in research process

Question 28

What is secondary data analysis?

Answer 28

If you are not using 2ndry data as you find it e.g. in literature reviews, or as an overview of topic under investigation, you may want to further analysis (secondary data analysis). The aim of secondary data analysis is to extract new findings and insights from the existing data. The factors affecting the quality / usefulness of the dataset for secondary data analysis are the same as those for the use of secondary data.

Question 29

What do you need to conduct secondary data analysis?

Answer 29

It is important to be in possession of a copy of the original questionnaire / discussion guide, description of sampling techniques used and source of data. May also be useful to have a copy of any instructions given to moderators or interviewers who conducted fieldwork, including any definitions or clarifications given to respondents. From a data point of view, it is useful to know: how the data were coded and analysed how the dataset is structured what technical tools were used in processing and analysis what weighting (if any) was applied list of variables and values and the coding and classification schemes used Non-response codes A list of publications produced from data will give better insight into the study.

Question 30

What is a data archive?

Answer 30

A data archive is a store / repository for data. Commercial organisations have their own data archives in the form of internal databases and data warehouses that form the basis of DSS’, or a management information system MIS e.g. external organisations may maintain data archives that can be accessed by anyone interested in using data collected there. A vast amount of data relating to social and economic life in the UK is held at the UK Data Archive. The archive contains data collected by the ONS on behalf of the UK government from regular, repeated surveys such as the Labour Force Survey, General Household Survey and Family Expenditure Survey.

Question 31

What is a data warehouse?

Answer 31

A data warehouse is a repository for data; in effect it is a very large database that contains data from one or more sources. It is a central storage facility that takes the concept of a data archive one step further, in that different datasets within the warehouse are integrated and elements in one set can be related to elements in another set (known as a relational database). Data that are stored in warehouses tend to be data that are useful for supporting management decision making within an organisation - in many cases this is the purpose of a data warehouse - to support the management-decision-making process. Such warehouses are often referred to as DSS or EIS or enterprise intelligence system (EIS). if a system is related to customers it may be referred to as a Customer Relationship Management System (CRM).

Question 32

What data is kept in a DSS or CRM?

Answer 32

The data in a DSS or CRM system tend to observational / behavioural data rather than survey research data. These observational data are drawn from customer interactions from EPOS scanners from payments and card transactions, from company’s website, from social media sites, mobile devices, electronic tags and chips e.g. RFIDs radio frequency identification, barcodes, labels, products. It may contain financial data or data related to performance management; it may be supply chain data. Essentially, it is data derived from any and all aspects in a business - thus it is important to bear in mind when scoping or planning a project as organization may hold data within DSS or CRM of use. You may not even need to conduct primary research.

Question 33

What is big data? How is it handled?

Answer 33

When data stored is such a size that it cannot be analysed and searched using standard computing techniques (when it reaches terabyte or exabyte size), it is referred to as ‘big data’. Information derived from a system is referred to as ‘analytics’ or ‘consumer analytics’ or ‘web analytics’ if concerned with web and website traffic. Analysis of big data relies on data minings with software that is referred to as analytics technology. Information produced may go towards customer insight, identifying emerging patterns and trends, assessing customer acquisition and retention, planning and advertising campaigns and other marketing activity, new product development.

Question 34

How are data warehouses designed or structured?

Answer 34

Data warehouse is designed or structured, and data given context in order to enhance decision support role and to make access to data in warehouse efficient and fast. There are two main designs or structures; relational database structure, based around star design with a central fact table for sales and several linked or related tables as arms of the star e.g. sales and product group / sales region / sales period… and the multidimensional database structured around a multidimensional cube design. The database or warehouse with traditional relational design has two main advantages over the MD design: it allows relatively easy integration of other relational databases and is easier to manage and update. Main advantage of MDDS over RDS is that it gives MD view of data.

Question 35

How is a data warehouse built?

Answer 35

Data is sent to the warehouse from what is called operational field e.g. scanner data of transactions from EPOS, RFID tags, companies website or social media site or from GIS. once in the warehouse they are referred to as being in the informational field. Data sent to the warehouse should be of good quality, as it will be used to support key management decisions and any inaccuracies or inconsistencies will impact on the quality of decision making. It is good practice to clean data before sending them to ensure that they are accurate and complete, that definitions of terms and variables and the coding procedures for these are consistent so that data can be fully integrated / fused with other data in the warehouse. Only relevant data to the needs of the DSS should be sent to the warehouse, unnecessary data will clog up system and slow access and processing time.

Question 36

How should you organise data in a warehouse?

Answer 36

End use of the warehouse should dictate structure and organisation of data. Should be done in a way that allows analysts to look at data from relevant perspectives e.g. customer types, by brand / market. Data should be stored in a way that facilitates faster access to more frequently used data but allows older data to be pulled up for reference. The warehouse contains information telling users about structure and how to find a way around it - this information is called metadata. It acts as a map of the warehouse, contents list including details of databases or sets in the warehouse, elements contained within them and how these relate to other databases in the warehouse.

Question 37

Key characteristics of a well-designed data warehouse?

Answer 37

Key characteristics of a well-designed data warehouse are: Ability to store ever-increasing volumes of data without affecting processing performance User friendly Everyone can access regardless of locations Many users can use at the same time without impacting processing speeds Facilitates analysis of data from a variety of perspectives

Question 38

What is data mining?

Answer 38

Databases and data warehouses can be very large - terabyte and exabyte size, with millions of rows and hundreds of variables. Data mining, also known as knowledge discovery in databases (KDD), is the process of mining / extracting information and knowledge that are useful to decision makers using automated techniques and parallel and multi parallel computer technology, cluster computing and large-scale grid computing techniques. A data mining program can manipulate data, combining variables, for example and allowing the user to select elements or sections of the database for analysis; provide basic descriptive statistics, look for associations and relationships between variables and perform cluster analysis. Where data mining differs from other data analysis techniques is the volume of dtaa it can process and analyse, and its ability to discover patterns and relationships that cannot be detected with standard analysis techniques. Data mining systems divide workload between a set of parallel processors, enabling streams of data to be processed simultaneously.

Question 39

Approaches to mining?

Answer 39

Verification approach - already have an idea about patterns of behaviour or relationships between variables - you are testing a formulated hypothesis. Discovery approach - you haven't got a clear idea about patterns and want to find out what exists amongst data by getting a data mining program to search and explore databases in order to find patterns and relationships. It will do this by getting to know data; learning rules that apply within the database; identifying how elements relate to each other; what networks exist in data.

Question 40

What is data integration?

Answer 40

Process of merging data from different sources. Uses techniques of data fusion and modelling to achieve this - end purpose is to build a more detailed picture / profile of the consumer than is provided by a single dataset. Data linking is the linking of data from different sources thought to belong to the same individual.