Data Collection Design PDF

Faculty of Management Department of Venture Nizhny Novgorod Creation Data Collection Design Seminar 4 Dr. Waqas Ahmed  2 Agenda Sampling Survey Measurement Scales Piloting Ethical Considerations ...

Faculty of Management Department of Venture Nizhny Novgorod Creation Data Collection Design Seminar 4 Dr. Waqas Ahmed  2 Agenda Sampling Survey Measurement Scales Piloting Ethical Considerations  3 SAMPLE The selection of a small number of elements from a larger defined target group of elements and expecting that the information gathered from the small group will allow judgments to be made about the larger group. A subset of the population, and that subset is called a sample. Population Sampling For several reasons, researchers typically unable to study the entire population. Sampling is an act, process or technique of selecting a suitable sample, or a representative part of population for the purpose of determining parameters or characteristics of whole  4 Sampling Techniques Non-Probability sampling Types of Nonprobability sampling Non-probability sampling is a sampling Convenience sampling method in which not all members of the Samples drawn at the convenience of the population have an equal chance of researcher. Assumption is that the defined participating in the study. The samples based target population is homogeneous, and the on the subjective judgment of the researcher samples are similar to overall target rather than random selection. population. Limited resources or no sampling frame, or Judgment sampling need to select a particular case because of Participants are selected according to the the nature of your research question researcher’s or some other experienced (particularly common with qualitative data) individual’s belief that they will meet the requirements of the study. Assumption is the researcher’s subjective belief that the opinions of a group of perceived experts on the topic of interest are representative of the entire defined target population  5 Sampling Techniques Types of Nonprobability sampling Types of Nonprobability sampling Quota sampling Snowball sampling Divide population → quota for each unit → Involves subjectively identifying and convenience sample for each unit until quota. qualifying a set of initial prospective Provides an assurance that pre-specified respondents who can, in turn, help the subgroups of the defined target population researcher identify additional people to be are represented on pertinent sampling included in the study. Referrals will have factors that are determined by the researcher demographic and psychographic characteristics that are more similar to a person referring than would be by chance.  6 Sampling Techniques Probability Sampling Types of Probability sampling Probability sampling Simple Random Sample Sampling in which each item in the population A sampling procedure that ensures that each has an equal chance (this chance is greater element in the population will have an equal than zero) for getting selected is called chance of being included in the sample probability sampling. Chance of each case/element being selected is known and usually equal. Probability Sampling uses lesser reliance over the human judgment which makes the overall process free from over biasness.  7 Sampling Techniques Types of Probability sampling Systematic Random Stratified Random Sample Cluster Sample Sample Subsamples are drawn within Subsamples are drawn within Sample in a systematic way, different strata using simple different strata using simple where every name from the random sampling. Each random sampling. Each list will be drawn. stratum is more or less equal stratum is more or less equal on some characteristic. on some characteristic.  8 Survey Research Distinguished by the need to collect data from large samples of people. Generally associated with descriptive and causal research. Administered to selected individuals with responses recorded in a structured and precise manner. Provide specific facts and estimates to: Make predictions about relationships. Understand the relationships and differences. Validate existing relationships.  9 Survey Method  10 Measurement Levels Nominal scale Ordinal scale Questions “assign” some sort of descriptor Respondent can express the relative as the magnitude between the answers to a response. response. (rank ordered) The response does not contain any kind of Data can be arranged in a “greater intensity or rank. than/less than” or “bigger than/smaller Responses can only be categorised into than” pattern. mutually Cannot determine the absolute difference exclusive subsets with no relative between the responses. magnitudes between them. Example: The order people finish a race Example: Are you male or female? (Male = (either 1st , 2nd or 3rd) 1; Female = 2)  11 Measurement Levels Interval scales Ratio scale Respondent can express the relative Respondent can express absolute magnitude between the responses and the differences between each scale point and absolute differences between each to make absolute comparisons between response. the scale points. Distance property allows for powerful Scale allows a true natural zero or statistical analysis. comparison point. Distances between each response does not The distances between each scale point have to be equal. can be compared based on a true zero Does not allow for absolute comparisons. point, thus allowing for absolute comparisons.  12 Responses: Nominal Data are neither measured nor ordered but subjects are merely allocated to distinct categories Example: Are you: 1. Male 2. Female What is your marital status:  Single  Married  Widowed  Divorced  Separated  Other  13 Responses: Nominal Limited choices of responses, lack of consistency in what a yes/no, agree/disagree response means Example: Do you have trouble training your employees? Attitudes and behaviors lie on a continuum To what extent do you experience difficulty when training your employees?  None  A little  Quite a bit  A lot  I do not train employees  14 Responses: Ordinal Data about the rank order of scores Example: What is the highest level of education you have reached: Lowest Education  Did not complete primary school  Completed primary school  Up to, but not including year 10  Completed year 10 or equivalent  HSD or equivalent  Associate Degree  BA or equivalent  Masters Highest Education  PhD or equivalent (JD, DBA, etc.)  15 Responses: Ordinal Continuous Types of ordinal continuous response scales – Visual analogue scales Example: Overall, how much satisfaction do you get from your job? Low satisfaction ______________ High satisfaction  16 Responses: Ordinal Continuous Provide adjectives for points along the line (adjectival scales) I No Sat I little Sat I Some Sat I high Sat I very high Sat I  17 Responses: Likert scale Rate agreement with a series of statements Example: To what extent do you agree or disagree with each of the following statements:  Strongly Agree  Agree  Neither  Disagree  Strongly disagree  18 Responses: Likert scale How many steps/boxes should there be? – five to seven response categories ideal People averse to extreme ends of a scale – avoid absolutes; almost always vs always, almost never vs never – add throw away categories at either end  19 Responses: Likert scale Should there be an even or odd number of categories – not an issue if your scale goes from ‘not at all’ to ‘very much’ (unipolar scales) – If your scale is bipolar (eg: strongly agree to strongly disagree), decide whether you want a ‘neutral’ point  20 Scales’ structure, statistics and transformation Mathematical Permissible Admissible Scale Scale Type structure Statistics Transformation Nominal Standard set structure One to One Mode, Chi-Squared categorical (unordered) (equality(=)) Monotonic increasing Ordinal Totally ordered set Median, percentile (order(