Chapter 5 Data Collection Methods (PPT slides)

Find here the PPT for Chapter 5 Data Dollection Methods

4.7 The importance of High response rate and the actual sample size required

The importance of a high response rate

The most important aspect of a probability sample is that it represents the population. A perfect representative sample is one that exactly represents the population from which it is taken. You, therefore, need to obtain as high a response rate as possible to ensure that your sample is representative. This requires you to determine the  minimum adjusted sample size , and finally the actual sample size required.

This note gives you a brief discussion on how to compute the adjusted sample size by considering the non responses , and also the  actual sample size you require for your research.
click here for the note. 

4.6 What sample size do I need for my research?

 What sample size do I need for my research?

 Perhaps the most frequently asked question concerning the determination of sample size is    “What should be my sample size?” The article by Isreal (2013) provides you a summary of the most important points that one needs top consider to answer such a question.

In fact ,you need to consider a number of factors to determine the appropriate sample size. These include , the purpose of the study, population size, the level of precision, the level of confidence or risk, and the degree of variability in the attributes being measured. For full discussion of these issues, Read the article below  DETERMINING SAMPLE SIZE. It discusses all these factors and how to use published tables, and applying formulas to calculate a sample size.

Article: Determining the sample size 

More precision, more sample size

The table below provides sample sizes for different sizes of population at a 95% confidence level and margin of error ranging from 5% to 1%. Note that the smaller the margin of error ( which means when we need more precise measurement of the variable ) , we need to have more sample size. 

Table: sample sizes for different margin of errors

4.5 Impact of factors to choose sampling techniques

Impact of factors to choose sampling techniques

The impacts of factors to choose probabilistic and non-probabilistic sampling techniques are tabulated below.

Table: 1  

Table: 2 

4.4 Comparison of probability sampling designs

Comparison of probability sampling designs

The table below depicts the comparison of probability sampling.

Table: 

Furthermore, a detail comparison between the cluster and stratified sampling is presented in the table below.

Table: 

4.3 Most frequently used sampling techniques

Most frequently used sampling techniques

The diagram below shows the most commonly used sampling techniques.

Diagram: 

The decision trees below helps you to make a decision about which  type of sampling techniques to use for your research.  

Decision Tree 1: 

Decision Tree 2:  

4.2 Characteristics of a good sample: accuracy and precision

Characteristics of a good sample: accuracy and precision

  

Two conditions are appropriate for a census study. These are when a census is (1) feasible     (meaning when the population is small),  and (2) necessary (meaning  when the elements are quite different from each other.  However , there are several compelling reasons for sampling, including (1) lower cost, (2) greater accuracy of results, (3) greater speed of data collection, and (4) availability of population elements.

What is a good sample then? The ultimate test of a sample design is how well it represents the characteristics of the population it purports to represent. In measurement terms, the sample must be valid. Validity of a sample depends on two considerations:  accuracy and precision.

Accuracy is the degree to which bias is absent from the sample. When the sample is drawn properly, the measure of behaviour, attitudes, or knowledge (or the measurement variables) of some sample elements will be less than (thus, underestimate) the measure of those same variables drawn from the population. Also, the measure of the behavior, attitudes, or knowledge of other sample elements will be more than the population values (thus, overestimate them). Variations in these sample values offset each other, resulting in a sample value that is close to the population value. For these offsetting effects to occur, however, there must be enough elements in the sample, and they must be drawn in a way that favors neither overestimation nor underestimation.

Therefore , an accurate (unbiased) sample is one in which the underestimators offset the overestimators. Systematic variance has been defined as “the variation in measures due to some known or unknown influences that ‘cause’ the scores to lean in one direction more than another.”

Precision is measured by the standard error of estimate, a type of standard deviation measurement; the smaller the standard error of estimate, the higher is the precision of the sample. The ideal sample design produces a small standard error of estimate. However, not all types of sample design provide estimates of precision, and samples of the same size can produce different amounts of error. 

4.1 Characteristics of a good sample: accuracy and precision

Characteristics of a good sample: accuracy and precision

  

Two conditions are appropriate for a census study. These are when a census is (1) feasible      (meaning when the population is small),  and (2) necessary (meaning  when the elements are quite different from each other.  However , there are several compelling reasons for sampling, including (1) lower cost, (2) greater accuracy of results, (3) greater speed of data collection, and (4) availability of population elements.

What is a good sample then? The ultimate test of a sample design is how well it represents the characteristics of the population it purports to represent. In measurement terms, the sample must be valid. Validity of a sample depends on two considerations:  accuracy and precision.

Accuracy is the degree to which bias is absent from the sample. When the sample is drawn properly, the measure of behavior, attitudes, or knowledge (or the measurement variables) of some sample elements will be less than (thus, underestimate) the measure of those same variables drawn from the population. Also, the measure of the behavior, attitudes, or knowledge of other sample elements will be more than the population values (thus, overestimate them). Variations in these sample values offset each other, resulting in a sample value that is close to the population value. For these offsetting effects to occur, however, there must be enough elements in the sample, and they must be drawn in a way that favors neither overestimation nor underestimation.

Therefore , an accurate (unbiased) sample is one in which the under estimators offset the over-estimators. Systematic variance has been defined as “the variation in measures due to some known or unknown influences that ‘cause’ the scores to lean in one direction more than another.”

Precision is measured by the standard error of estimate, a type of standard deviation measurement; the smaller the standard error of estimate, the higher is the precision of the sample. The ideal sample design produces a small standard error of estimate. However, not all types of sample design provide estimates of precision, and samples of the same size can produce different amounts of error. 

Chapter 4 SAMPLING PPT slides

 Go to this link Chapter 4 SAMPLING PPT slides  to obtain the power point slides for chapter 4.

3.6 Notes on KR-20 and Cronbach Alpha

3.6 Notes on KR-20  and Cronbach Alpha 
You may find the following brief notes useful in understanding the essences of measuring the internal consistency of your data   collection tools. Use the following links to read the notes 
Note 1 
Note 2 
  

3.5 Reliability tests

3.5 Reliability tests

Consider the following reliability tests:

·       Test –retest

·       Split half method

·       KR 20

·       Cronbach alpha

a)Which one of them measure internal consistency? Which one them measure stability of a data collection tool like a  questionnaire?

b)Discuss their main features of these reliability tests. 

3.4 External and internal validity

3.4 External and internal Validity  

The two major forms of validity are the external and internal validity. The external validity of research findings is the data’s ability to be generalized across persons, settings, and times  

The Internal validity is about the ability of the research instrument to measure what it is purported to measure. It answers the question “does the instrument really measure what its designer claims it does?".  It consists of the following major forms: content validity, criterion-related validity, and construct validity, among others.

Discuss the methods that you could use to measure the different types of internal validity of a data collection instrument?    

3.3 Measurement scales

3.3  a) Provide  examples from business studies to illustrate differences among the following types of measurement  scales: nominal , ordinal, interval and ratio scales.

  

b) Provide examples how some constructs or variables could   be designed by a researcher as a categorical or a continuous variables.

3.2 Types of variables II

3.2 Provide examples of Research titles that help illustrate the difference among the following types of variables: 
Dependent variable  , Independent  variable, Moderating variable, Intervening (or mediating) variable, Control variable, and Confounding variable.

3.1 Types of variables I

3.1 The following schematic diagram depicts the various types of variables that could be considered in a research that aims at investigating the "Introduction of a four-day working week on workers' productivity".
(Note:  that it is not always a must to describe or list all types of variables in a research title, particularly as there may be infinite extraneous variables.) 

2.4 Research design and sampling design

2.4 When developing a business research proposal , discuss what you may include in the research design , and sampling design sections. How important is it to distinguish between these two sections?

2.3 Softwares for data analysis

2.3   List five application softwares available to you as researcher that may aid  your qualitative and quantitative data analysis. Discuss your preferred choice(s) and share their main features with your colleagues.

2.2 Important elements of a business research proposal

        2.2  There are a number of templates suggested by  different  authors when developing a business research proposal. Consider yourself as a manager who sponsors a business research or one who actual develop the proposal.What are the most important elements you would like to see included in the business research proposal? Why?

2.1 Distinguishing the "true" business problem from its symptoms

2.1 Discuss the challenges that you as a researcher may face to isolate the true business problem from its symptoms. What steps could you take in order to define your research problem as clearly as possible? Provide practical examples to illustrate the process.

chapter 3 MEASUREMENT CONCEPT IN BUSINESS RESEARCH PPT slides ( Nov 8)

Find  here the PPT slides for Chapter 3: Measurement Concept in Business Research

1.4 Basic Vs Applied research

1.4    The following table depicts the comparison between basic and applied type of researches

Basic vs. applied research continuum

Basic  research                                                                                                                 Applied research
Purpose	Purpose
·       Expand knowledge of processes of business and management ·       Results in universal principles relating to the process and its relationship to outcomes ·       Findings of significance and value to society in general	·       Improve understanding of particular business or management problem ·       Results in solution to problem ·       New knowledge limited to problem ·        Findings of practical relevance and value to manager(s) in organisation(s)
Context:	Context:
·       Undertaken by people based in universities ·       Choice of topic and objectives determined by the researcher ·        Flexible time scales	· Undertaken by people based in a variety of settings including organisations and universities · Objectives negotiated with originator ·  Tight time scales

1.3 Ethics in Business Research ( Nov 2015)

·        1.3  Here is a checklist to help you anticipate and deal with business research ethical issues. ( Saunders et al. , 2009)

      ✔  Attempt to recognize potential ethical issues that will affect your proposed research.

✔ Utilise your university’s code on research ethics to guide the design and conduct of your research.

✔ Anticipate ethical issues at the design stage of your research and discuss how you will seek to control these in your research proposal.

✔ Seek informed consent through the use of openness and honesty, rather than using deception.

✔ Do not exaggerate the likely benefits of your research for participating organisations or individuals.

✔ Respect others’ rights to privacy at all stages of your research project.

✔ Maintain objectivity and quality in relation to the processes you use to collect data.

✔ Recognise that the nature of an interview-based approach to research will mean that there is greater scope for ethical issues to arise, and seek to avoid the particular problems related to interviews and observation.

✔ Avoid referring to data gained from a particular participant when talking to others, where this would allow the individual to be identified with potentially harmful consequences to that person.

✔ Covert research should be considered only where reactivity is likely to be a significant issue or where access is denied (and a covert presence is practical). However, other ethical aspects of your research should still be respected when using this approach.

✔ Maintain your objectivity during the stages of analysing and reporting your research.

✔ Maintain the assurances that you gave to participating organisations with regard to confidentiality of the data obtained and their organisational anonymity.

✔ Consider the implications of using the Internet and email carefully in relation to the maintenance of  confidentiality and anonymity of your research participants and their data, before using this means to collect any data.

✔ Protect individual participants by taking great care to ensure their anonymity in relation to anything that you refer to in your project report unless you have their explicit permission to do otherwise.

✔ Consider how the collective interests of your research participants may be adversely affected by the nature of the data that you are proposing to collect, and alter the nature of your research question and objectives where this possibility is likely. Alternatively, declare this possibility to those whom you wish to participate in your proposed research.

✔ Consider how you will use secondary data in order to protect the identities of those who contributed to its collection or who are named within it.

✔ Unless necessary, base your research on genuinely anonymised data. Where it is necessary to process

personal data, comply with all of the data protection legal requirements carefully.

✔ Utilise your university’s code on research ethics to guide the design and conduct of your research.

✔ Anticipate ethical issues at the design stage of your research and discuss how you will seek to control these in your research proposal.

✔ Seek informed consent through the use of openness and honesty, rather than using deception.

✔ Do not exaggerate the likely benefits of your research for participating organisations or individuals.

✔ Respect others’ rights to privacy at all stages of your research project.

✔ Maintain objectivity and quality in relation to the processes you use to collect data.

✔ Recognise that the nature of an interview-based approach to research will mean that there is greater scope for ethical issues to arise, and seek to avoid the particular problems related to interviews and observation.

✔ Avoid referring to data gained from a particular participant when talking to others, where this would allow the individual to be identified with potentially harmful consequences to that person.

✔ Covert research should be considered only where reactivity is likely to be a significant issue or where access is denied (and a covert presence is practical). However, other ethical aspects of your research should still be respected when using this approach.

✔ Maintain your objectivity during the stages of analyzing and reporting your research.

✔ Maintain the assurances that you gave to participating organisations with regard to confidentiality of the data obtained and their organisational anonymity.

✔ Consider the implications of using the Internet and email carefully in relation to the maintenance of  confidentiality and anonymity of your research participants and their data, before using this means to collect any data.

✔ Protect individual participants by taking great care to ensure their anonymity in relation to anything that you refer to in your project report unless you have their explicit permission to do otherwise.

✔ Consider how the collective interests of your research participants may be adversely affected by the nature of the data that you are proposing to collect, and alter the nature of your research question and objectives where this possibility is likely. Alternatively, declare this possibility to those whom you wish to participate in your proposed research.

✔ Consider how you will use secondary data in order to protect the identities of those who contributed to its collection or who are named within it.

✔ Unless necessary, base your research on genuinely anonymised data. Where it is necessary to process personal data, comply with all of the data protection legal requirements carefully.

1.2 Exploratory , descriptive , and explanatory (casual ) researches ( Nov 2 ,2015)

1.2   Provide examples for “exploratory , descriptive , and explanatory (casual ) type of business researches from your organizational context or others.  

1.1 Research as systematic, cyclic, and iterative Process ( Nov 2 ,2015)

  1.1 Why is business research  characterized as  a “ systematic,       
       cyclic, and iterative Process” ?