The data those have been collected already and readily available from other sources are called as secondary data. When compared to primary data, these secondary data are cheaper and more quickly obtainable. Usually, desk-based research is used to collect secondary data. After arriving at the secondary data, the researcher should examine the validity and reliability. Thus, the researcher should consider the secondary data which is highly valid and well-referenced in academic articles (Creswell, 2003).
Types of secondary data
Secondary data is categorised into internal (from the organization which is under observation) data which is routinely supplied by management and external (from outside the organization) data which is obtained from various resources such as internet, journals, books, directories, Non-governmental statistical data and Census data.
Figure 1. Types of Secondary data (Source: https://www.ilo.org)
Based on the suggestions of Bryman (1989); Dale et al., (1988); Hakim, (1982 ) the secondary data is also classified into documentary data, survey based data and multiple source secondary data.
Written documents like notices, correspondence, minutes of meetings, reports to shareholders, diaries, transcripts of speeches and administrative and public records will come under this category. Non-written documents such as Tape and video recordings, pictures, drawings, films and television programmes, DVD/CD can also be considered as documentary data.
Survey-based secondary data
The data collected by questionnaires that have already been analyzed for their original purpose are called as survey-based secondary data. They can be from:
- Continuous/regular surveys
- Ad hoc surveys
Multiple-source secondary data
Multiple-source secondary data can be from either documentary or on survey data, or it could be the combination of these data. These data can be used in cohort studies those have been done for the same population over a period of time and also to develop area-based data sets.
Sources of secondary data
Reports from central, state or local government
Reports from international firms and foreign governments
Journals, magazines and books
Publications from research scholars, universities and research groups- are the major sources of secondary data (Dash, 2011).
Uses of secondary data
Although all data is intended to provide information for analysis and decision making, secondary data can be used in several ways in the context and conduct of a research/consultancy project. In accordance to Malhotra and Birks (2000) and McDaniel and Gates (2004) secondary data can be useful to
- To identify the research problem
- To develop a strategy to arrive at the solutions for the research problem
- To construct a sampling plan
- To formulate a suitable research design
- To find out the answers for certain research questions or to test some hypotheses
- To interpret primary data
- To validate the outcomes from qualitative research
- To identify the potential problems
- To obtain the required background information and to improve the credibility of the study.
Evaluation of secondary data
The researcher has to view the secondary data with the same caution as he does with any primary data. He has to check whether he could access the data and the available secondary data could support him to arrive at the research objectives. Based on the following criteria the researcher should evaluate the secondary data resources (Stewart, 1984).
It is necessary to evaluate the secondary data based on the methodology through which it has been collected (Stewart, 1984). Hence, the researcher has to evaluate the factors like sampling procedure, size of sample, rate of response, field work procedures and data analysis methods.
Accuracy of the data from the secondary resource should be assessed by the researcher in order to make sure about the trustworthiness of his study. However, the specifications as well as the methodologies won’t be given in detail to the researcher. As a result, it is often very tough to assess the accuracy of secondary data. However, the researcher could check the accuracy through triangulation research.
Date of data collection
As the secondary data is associated with the events those have already happened, it is usually outdated. Hence the researcher has to view the date of data collection, duration between data collection and publication and the relevancy of the data with respect to existing situations. In the case of census data, date of data collection is the major issue as it is collected once in few years.
Purpose of data collection
Again, by definition, secondary data is data that was collected with some other purpose or objective than that which the researcher/consultant now addressing this data is concerned with. The researcher must assess the extent to which data collected with another purpose or specific objective in mind is appropriate and relevant to the researcher’s situation or problem.
Nature: content of data
The data with high validity and accuracy can’t be useful if the content of the data is not good. Sometimes, there may be no proper link between the relationships examined by the data and the measurement categories adopted by the researcher. For example, the data may consider the relationship between salary levels and motivation. However, motivation may have been defined and measured in a way that is inconsistent or inappropriate with how the researcher considering the secondary data wishes to measure motivation.
It is suggested that the factors such as expertise, credibility and overall trustworthiness of the source should be considered by the researcher while evaluating the secondary data resources. In general, Government reports are found to be more credible than other commercial sources of secondary data. If the researcher knows the provider of the data or the primary data collection methodologies adopted by the researcher or the source of the primary data (whether original or acquired source), dependability and source credibility will become high (Stewart ,1984).
These are the key factors associated with the evaluation of secondary data resources. In addition to these factors other features such as costs and benefits, issues in access as well as control over data quality will also be considered while evaluating the secondary data (Stewart & Kamins, 1993; Denscombe, 1998).
Advantages of secondary data
Researchers (Boslaugh, 2007) have observed the following advantages with secondary data:
- Less resource requirements
- Inconspicuous method
- Applicable for longitudinal studies
- Comparative and contextual data can be obtained
- Lead to unforeseen discoveries
- Durability of data
Disadvantages of secondary data
- The purpose for which the data is collected may not match the objectives of the researcher. Sometimes, it may be very difficult to get the access to the data
- Aggregations and definitions may not match the expectations of the researcher.
- The researcher has no control over data quality
Using more than one method
Sometimes, the researcher may be in a situation to adopt more than one method of data collection. For example, he may conduct secondary research followed by observation and focus group interviews. As this approach is the combination of two or more methods it is referred to as triangulation. Most of the management and consultancy research are not exclusive, and dichotomous. They may include both qualitative and quantitative research. On the other hand, focus groups and surveys can be done together for experimental research.
Factors affecting the selection of the research method
The research methodology adopted by the researcher has extreme influence over the success of the study. Hence, he should be very careful while selecting the research methodology. To arrive at the outcomes with expected quality, he should consider various factors while selecting the research methods. Among them, validity, reliability, and generalizability are the important factors that influence the selection of research method.
Validity and reliability of measures
The two important parts of research process are validity and reliability of measures. As stated by (Bryman & Cramer, 2005) “It is generally accepted that when a concept has been operationally defined, in that a measure of it has been proposed, the ensuring measurement device should be both valid and reliable”. Accuracy or exactness of data is related to the validity measure (Churchill & Lacobucci, 2002) which denotes the accuracy of the survey instrument (Litwin, 1995). As stated by Oppenheim(1984), Baines and Chansarkar (2002), Parasuraman, (1991), Peterson, (2000), validity measure could check whether the survey instrument measures the variables under study. Validity can be inferred through direct assessment employing validity and indirect assessment via reliability (Churchill & Lacobucci , 2002).
Generalizability, another dimension of validity is related to the degree with which the research outcomes could be generalised to other situations. Generalizability of data is crucial in two aspects. (i) Sampling is often used in the generation of data in the process of research and consultancy projects. The researcher should be able, to examine the extent to which results from the sample will also be present in the wider population from which the sample is obtained. Thus, Generalizability, is not associated only with the data collection methods but also with sample design and sampling method. (ii) The extent to which the data and results of a particular research project can be generalized to other situations. This, of course, is crucial in developing theories and particularly in the deductive approach to research.
Hence, it is understood that these three measures are found to support the researcher to generate scientific and reproducible research outcomes. It was believed that these three measures were developed for conducting research in the field of physical sciences where quantitative data is mostly used (Easterby-Smith et al., 2002).
Objectives/purpose of research
The objectives and purpose of the research extremely influences the selection of the research method. For example, the researcher may attempt to arrive at the solutions for a particular set of questions in his study. Hence, he could choose the methodology or techniques based on these objectives.
Skills and expertise of the researcher
It is true that an experienced researcher could opt for any kind of alternative research methodology and data-collection techniques. But each and every researcher can’t be an expert in all types of research methods. The researcher should opt for the methodology in which he has the high degree of expertise and familiarity. This factor is the major determinant for the success of the study/research.
The cost of the proposed research methodology is an important factor in determining the choice of research methods. Mostly, cost of the data collection method is high in many projects. Sometimes, the most effective research methodology in quality and potential value may be rejected in favour of a slightly less effective research methodology which is cheaper.
The researcher should consider the time given to complete the research while choosing the research methodology. If he has allotted with short duration, he should not opt for time consuming data collection methods such as large-scale surveys.
To make the study easier, the researcher should opt for readily available and accessible research methods. As the secondary data is already available, it is very easy to collect them. Sometimes, the data may not be available. In this case it is necessary to choose an alternative method to collect data.
Preferences/values of the consultant
The consultant should not allow his/her preferences and values to influence the choice of research methodology and he/she should give importance for the preferences of the client and the validity and cost of the methods.
Preferences/values of the client
Preferences and values of the client are found to extremely influence the choice of alternative research methodologies in the case of consultancy research projects. For example, the consultancy may feel that data for a particular project can be effectively collected through observation. But the client may tell that his department won’t allow observation method. In this case, it is necessary to discuss in terms of research methodology at the contracting stage of the consultancy process.
Ethical, legal and other issues
The consultant and client should consider and take care of several issues while selecting alternative research methodologies. Ethical issues concerned with the choice of research methodology and data collection techniques are one of these issues. Sometimes, there may be some issues in informing the participants that they are under observation (Wysocki 2004). On the other hand, there may be some issues in terms of collection, storage and uses of data. Most of the countries deal these issues with the help of legislation. For example, Data Protection Act has been passed in the UK to deal these issues.
Some of the factors should consider before utilizing secondary data
At Statswork, we are very conscious in collecting the secondary data include
- 1. Checking the reliability of data
- 2. Accuracy and quality of data for specific research goal (level of accuracy and quality)
- 3. Suitability of data
How do I contact Statswork expert for secondary data collection?
You can reach us at any time in a week (between Monday to Saturday, 09.00 am to 06.00 pm IST) at +91 87544 46690 or fill up this form or Just mail us via LinkedIn.