Wednesday, April 18, 2007

Multiphase Sampling

  • Definition:
    A sampling procedure in which some information is collected from the whole sample and additional information is collected, at the same time or later, from sub samples of the entire sample (i.e. some units provide more information than others).
  • Description:
    A multi-phase sample collects basic information from a large sample of units and then, for a sub sample of these units, collects more detailed information. The most common form of multi-phase sampling is two-phase sampling (or double sampling), but three or more phases are also possible.

    Multi-phase sampling is useful when the frame lacks auxiliary information that could be used to stratify the population or to screen out part of the population.

    Example 1: Suppose that an organization needs information about cattle farmers in Alberta, but the survey frame lists all types of farms—cattle, dairy, grain, hog, poultry and produce. To complicate matters, the survey frame does not provide any auxiliary information for the farms listed there.A simple survey could be conducted whose only question is "Is part or all of your farm devoted to cattle farming?" With only one question, this survey should have a low cost per interview (especially if done by telephone) and, consequently, the organization should be able to draw a large sample. Once the first sample has been drawn, a second, smaller sample can be extracted from among the cattle farmers and more detailed questions asked of these farmers. Using this method, the organization avoids the expense of surveying units that are not in this specific scope (i.e., non-cattle farmers).

    Multi-phase sampling can be used when there is insufficient budget to collect information from the whole sample, or when doing so would create excessive burden on the respondent, or even when there are very different costs of collection for different questions on a survey.

    Example 2: A health survey asks participants some basic questions about their diet, smoking habits, exercise routines and alcohol consumption. In addition, the survey requires that respondents subject themselves to some direct physical tests, such as running on a treadmill or having their blood pressure and cholesterol levels measured.Filling out questionnaires or interviewing participants are relatively inexpensive procedures, but the medical tests require the supervision and assistance of a trained health practitioner, as well as the use of an equipped laboratory, both of which can be quite costly. The best way to conduct this survey would be to use a two-phase sample approach. In the first phase, the interviews are performed on an appropriately sized sample. From this sample, a smaller sample is drawn. This second sample will take part in the medical tests.
  • Difference & Similarities between Multiphase & Multistage Sampling:
    Multi-phase sampling is quite different from multi-stage sampling, despite the similarities in name. Although multi-phase sampling also involves taking two or more samples, in multiphase sampling all samples are drawn from the same frame and at each phase the units are structurally the same. However, as with multi-stage sampling, the more phases used, the more complex the sample design and estimation will become.

  • Merits:
    It leads to inferences of predetermined precisions based on a number of observations.
    It reduces the error.

  • Demerits:
    It is time consuming & costly
    It's planning & administration is complicated

    Activity: Now since you'll know what multiphase sampling is, can you think of a merit & a demrit other than mentioned above. Write them down in the comments with proper rationale.

Sunday, April 15, 2007

Randomeness has types ??... thinking...

This post is with reference to the post by Hemangi 'Is randomeness only of one type?' ref.

Thedifference between the two examples:
Example 1:
  • The sample is selected from sample frame randomly
  • Survey is conducted

Example 2:

  • The sample is not selected from sample frame randomly. But it is selected from a particular group from the sample frame randomly.
  • Experiment is conducted by forming 2 groups which are tested on two different independet variables.

Sunday, April 8, 2007

Simple Random Sampling vs Convinience Sampling

This post is with reference to the post 'lets get sampling' by Hemangi.
Ref. -

Simple Random Sampling:
  • Each individual/study unit from the sample frame has equal chance of getting selected
  • The results will be applicable to entire population(sample frame)
  • Complete list of all the study units in the sample frame is required
  • Representativeness of the sample is given more importance
  • Kown as lottery method

Convinience Sampling:

  • Each individual/study unit from the sample frame dose not have the equal chance of getting selected. Only easily accessible by the researcher have the chance of getting selected
  • The results will be applicable to only the selected group of individuals
  • Complete list of all the study units in the sample frame may not be required
  • Accessibility of the sample is given more importance
  • Known as accidental sampling & opportunity sampling

Sunday, April 1, 2007

Criteria of sampling design- GO Measure Practical Economy!

This is with reference to by Hemagi.
  • Pic. 1 - timing as criteria , may be related to practicality
  • Pic. 2 - selecting individuals with particular characteristics, could be goal orientation

Hemangi , Im not sure about what is expected, & if Im right, please let me know,thanks.

Accessibilty of the sample

This post is with reference to post 2 by Hemangi.
The 3 examples of problems that researcher might face in accessing the sample are:
  • Absentism i.e. the selected individuals may remain absent for testing, questioning etc. E.g small children, ill individuals.
  • Pemission to access the selected individuals may be denied by authority. E.g. to prisoners
  • Sample itself may not allow for the access, may be due to lack of time or involvement of some sensitive issue e.g. Executives may not have time for your research, or victims of child abuse may be unwilling to share or disclose anything to anunknown researcher.

Friday, March 30, 2007


Hi ! This post is with regard to question asked in 'sampling' post by Hemangi. Ref.
According to me its very very difficult(or impossible) for a single researcher to do this task unles & until he/she is provided with lots of assistance.
Another thing is that even with the assistance it wont be easy to do so, as we all know how large population we have here in India! Also what amount of time, energy & other resources it will take to colllect & analyse the data....
So I think, this task is possible when provided with assistance as well as random sampling is done in each of the states & then results are generalised. As in example given by the Hemangi about books, we generalise the quality of few stories which are randomly picked up to entire book, the results of randomly picked population from each state(proportionate to that state's population) can be generalised to the entire country.