Page 165 - Data Science class 11
P. 165

Example
            Suppose you have selected a group of different types of people randomly standing in rows of 10 people each. Also,
            you have decided to survey every fourth person in a row as the population is very large. So, if you randomly select
            2nd person from first row, the other people you would select from the same row would be at positions 5th and 8th
            and so on for other rows.

            Advantages
            Following are the advantages of systematic sampling:
               • Evenly distributed samples: In systematic sampling, each individual is selected at a fixed distance, creating evenly
              distribution of collection of subjects. Since it is highly structured, it produces a more authentic representation of the
              entire population. As such, the results are easier to analyse.

               • Simplicity: Researchers are able to build, analyse and manage such samples easily because of its basic structure. The
              formula to select sample subsets is known prior, as such choosing the initial subject randomly is what is required.
              Thereafter, the selection process follows a fixed pattern, until the desired sample group is complete.
               • Unbiased: With systematic sampling, each participant is at a fixed distance from each other. Since there is a fixed
              interval, researchers have no control over which individuals are chosen for sampling. This helps reduce the chances
              for bias, favoritism, errors, and data manipulation.
            Disadvantages

            Following are the disadvantages of systematic sampling:
               • Possibility of uneven selection: Since systematic sampling draws conclusions from a subset of a population, results
              might not be completely precise. Systematic sampling depends on a numbering system to select sample participants.
              Responses of some participants are not included, so the results can’t be complete. Therefore, the researchers will
              always miss feedback, leading to a new finding.
               • Prediction of patterns: If the population being surveyed is small, the integer pattern used to choose samples can be
              easily predicted. This can cause bias and lead to erroneous responses among the participants.
               • Dependability of outcome on population count: It is the initial count of the population on which the outcome
              is dependent. After all, that’s the number that is divided by the desired sample size to determine the fixed interval
              for selecting sample. When the population isn’t measurable or available, a close approximation is required. If the
              population is smaller or larger than its actual number, this can produce wrong results.

            Convenience Sampling
            As the name suggests, convenience sampling is a type of non-probability sampling that involves the sample being
            drawn from that part of the population that is easier to find. In convenience sampling, people are sampled simply
            because they are 'convenient' sources of data for researchers. It's a first come, first serve sample.


                                                                                              Randomisation    163
   160   161   162   163   164   165   166   167   168   169   170