Research Population

The research population has some common characteristics that bind them together.
There are various sources from which a sample is created. A set of all these sources is called a Sampling frame from which the sample is selected. With the help of sampling frames, researchers are able to select the sample population from the target population that will be tested in the research or survey (StatPac Inc., 2014). A sampling frame should have the following characteristics:

Completeness: A frame should be complete in a sense that all the individuals of the target population should be covered in the frame. In case we are missing some of the individuals, the result will not be accurate and it will become pointless to waste time and resources on that research. For example, a survey is to be conducted on the number of children born in a particular area in a particular time period and we are counting only the ones which are born in some medical facilities. Then there can be chances that some are born at home or somewhere else but not a medical facility. So in such a case, these children will not be part of the sample and hence the results will be different from what it should be.

Also, there can be another kind of inadequate coverage, which poses problems for the surveys in which all the target population is considered from which the sample is to be taken, but some of the populatio
n may not be in town at that time of survey and so the data collected will be inadequate. For example, if some country wants to have a national survey of all the house holds, but the team doing it is not considering the ones who are living in some institutions, boat houses or nomads. So, more frames are required to cover these people also inorder to get proper data for analysis.

Accuracy: “to err is to human”. It is a well known saying that humans do mistakes and it’s a very natural tendency. We want our frame to be accurate, based on which our whole survey is based. Both trends have shown that mistakes are more in frames than in survey. A frame should include all the elements of the target population and there should be no repetition. For example, if we are listing all the organizations with 100 + employees, than any entry with 99 employees will be wrong. Also, all the organizations with 100+ should be in list without missing anyone (Royal Geographical Society, 2016).

Current frame: This characteristic is very important for a frame because only then a frame is able to fulfill the characteristics of accuracy and completeness. In case a frame is obsolete, that means that the data will be out dated and the information gathered from it is obsolete as well. The best example of such a frame is one done for population census. There is every probability that new residents have come due to new construction or may be some people have migrated to other places. So, the data can give no useful information on the current population of the place.

Categories of frames used in sampling

Area frames: An area sampling frame, country’s geographical units are arranged hierarchically. The hierarchy can be in a order like country, state, district, village, ward, block etc. They can also be further subdivided as area leaders (Kruse, 2009). For the purpose of the survey, some important characteristics required are:

• The entire land of the area where survey will be conducted should be covered.

• Boundaries should be well defined.

• Some population figures should be available for the research.

At times when national survey is done, coverage of the total land of the nation’s geographic area is vital. Only then researchers will be able to do authentic surveys based on it. Also, well marked boundaries will bring immense help to create genuine data (Buskirk, 2016). Whenever an area frame is to be generated, the starting is from the household surveys that leads to the nation’s population census. EA’s are formed to more subdivide the area for the convenience of the survey takers

An area frame is nothing but a list of all the elements of that area under survey. Because, it starts with making a list of different units in a population which are required in case of survey samples.

List frames. A list frame, as its name suggests has a list of the research population. In every country, there is a list of household surveys which is made when a census is taken. It contains a list of all the households of that country which are arranged geographically as discussed above .

When a recent census is done, it is very accurate and updated. It is so because it has all the properties of accuracy, completeness and updated. Since, it is arranged geographically, its stratification is very simple, if survey is to be conducted in any geographical area of the nation. In case some survey is to be conducted, then it’s the best time to take a sample and conduct a survey because the data is fresh and so, the research done will be very accurate and bring accurate results.
Multiple frames: Till now, we have discussed many sampling frames, which are used on target population for various purposes. Sometimes, we can use them in combination, usually two of them or it can be more than two also.

Typical dual frame in household surveys: Just to make it simple and easy to understand, we will discuss it by the principals of multi frame designs. In this, the area population of general frame is combined with the list frame of individuals who are part of the target population on which survey is done. For instance, in an unemployment survey, which is based on an area frame of households along with the list frame that contains people who are currently unemployed and have registered their names in ministry (The Regents of the University of Michigan, 2010).

According to me, to get a sample for clean edge razor, our company should go to area frames because then we will have the list of males and females of all the parts of the area where we want to release our product. Then, we can take out the list of males from it. Also, this is based on different areas and our product is for fastidious groomers which look for superior products. So, if they look for such products means they can afford it and so, the areas which are expensive to live can be targeted. The list of areas that can be taken from area frames.
Task 2
Main Sampling approaches can be classified into one of the two categories:
 Probability Sampling: Many times when we are short of time and money to work on everything in the target population, we randomly select some to work on. The selection may vary on the number of total population or some other factors. The selection can be a name from the draw, particular numbers, etc. So, any of such selection methods which are based on selecting randomly is called a probability sampling.

 Non-probability Sampling: As the name suggests, we are not relying on probability to get the sample from the research population. That means we are not selecting randomly unlike the probability sampling method. So, there can be a possibility that a non probability sample may not properly represent a population for which we are quite sure in probability sampling. Also, we are not able to understand in this that how well we have gathered results by testing this sample because it may or may not represent it properly. Most of the time if possible, researchers go for probability sampling, but sometimes due to lack of time, money or some other resources, it is not feasible to go for probability sampling (The Pennsylvania State University, 2016). Then, the researchers go for this sampling. It can be sub divided into two categories: purposive and accidental.

In our case, we should go for probability sampling, since we cannot target everyone in a population which will take lots of time and money. It covers a lot of area of UK, so after taking samples from an area frame of the target customers, we can do probability sampling on it. Also, we cannot wait so much inorder to test each and everyone since we have deadlines to meet and release this product. We will select some people randomly and test them. It has many types like simple random sampling, cluster sampling, stratified and systematic (Polit & Hungler, 2015).

Probability Sampling
In probability sampling, it take some random selection to do the survey. Since we are taking a random selection, so there has to be some assurance or methods by which we should get an equal probability of different parameters to be chosen. In this approach, we can know that which sampling units are from which sample and can be assured that each of them are considered.

The following sampling methods are examples of probability sampling:

1. Simple Random Sampling (SRS): This one is the easiest of all the samples methods and easy to understand and do. Other methods are basically based on it, but with some variations. In this method, a researcher takes elements from the sampling frame randomly. It is based on the probability theory according to which all the elements will have a chance to express in this way. The elements are to be numbered first and a list is made. Then, the researcher will select the numbers which are in mathematical order randomly. He can use some table numbers or many statistical books have numbers available. There are many computer programs available that can generate a random list.
There is no guarantee that every sample made from it will be perfect to show the population. But, we can infer that the sample represented will be close to target population and we can find out the probability of it.


2. Cluster Sampling: In this kind of sampling, the elements which are heterogeneous in nature as per the characteristics of group are taken. This will make a cluster of heterogenous elements and the homogeneous elements get clustered in other groups. In this way, clusters or groups are formed with similar characteristics. The main benefit of this sampling is that it is not very expensive. So, this sampling is different from another in a way that we are not choosing randomly from a population but making clusters first. Once these clusters are made, then sampling can be done from these clusters and then the test can be done to the individuals taken from these clusters. It helps in two ways: the cost is not high and so, it is more economical and in case of a dispersed population, if the researcher is not having an appropriate sampling frame, then this sampling helps a lot (Jewel, 2012).

3. Stratified Sampling: In case the target population on which the research is being conducted is heterogenous, then random selection will not produce an appropriate sample to test. Something may come in excess and some elements may lag behind and so, the theorem of probability will not work on this. In such cases, we have to do the stratification of the target population so as to make it homogenous and then to take out a sample of it by random sampling. So, from each mini stratums that are made in the target population are used to take a sub sample, out of which a final sample is drawn out (Gulf War Veterans, 2012). This sampling helps to increase efficiency and adequate & appropriate data because we are making sub populations before taking final sample.

4. Systematic Sampling : Systematic sampling provides a shortcut for selecting randomly and so requires less time. First of all, we have to number all the elements in the sample. The researcher will not use a list in which they have selected random numbers, but instead of that, researcher will decide a interval for sampling. For example 1, element in x where x will be the interval number (Turner, 2003). This way systematically, he can select a sample from a frame. It is also easy for doing calculations on it later.

Out of the methods mentioned above, I think we should go for stratified sampling due to following reasons: -
1. Stratified sampling is more efficient than the other probability sampling methods. Since, it makes different stratums which are homogenous among themselves in a group and differences with other groups. In our case also, not all males can afford the product or some may look for a particular thing in skin care or some may alter. So, to know the needs of different groups and then taking sample will bring more accurate results. It gives a big improvement in efficiency of sampling.
2. It gives a surety that the data of the sample is accurate because the criteria that we used to get it is very reliable.

3. Also, its feasible when the analyser wishes to study more qualities of a particular populations or subgroups. Therefore, if one wishes to conclude about activities in the different context of student body, stratified sampling will be in use

4. Also, it is used wherein different methods of data collection are applied in different parts of the population. This can be significantly found when we actually survey different organization’s employees using one method, but use a unique approach with employees distributed over the entire country

To implement this plan, firstly we will take a household list of area that we need to target first. That will become our area frame which we will use for all our further sampling procedures. Then, we will take out the list of males from that list since they are only end users of our product. Now, we will do probability sampling on it, because the area is so large that we cannot go to each and everyone to test them.

In probability sampling, we will do stratified sampling because Stratified sampling is more efficient than the other probability sampling methods. Since, it makes different stratums which are homogenous among themselves in a group and differences with other groups. In our case also, not all males can afford the product or some may look for a particular thing in skin care or some may alter. So, to know the needs of different groups and then taking sample will bring more accurate results. It gives a big improvement in efficiency of sampling. It gives a surety that the data of the sample is accurate because the criteria that we used to get it is very reliable and also its feasible when the analyzer wishes to study more qualities of a particular population or subgroups (StatPac Inc., 2014). Therefore, if one wishes to conclude about activities in the different context of the student body, stratified sampling will be in use. It is used wherein different methods of data collection are applied in different parts of the population. This can be significantly found when we actually survey different organization’s employees using one method, but use a unique approach with employees distributed over the entire country

Then a survey will be conducted on the sample of the target population which will test them on all our queries regarding its market, response, launch. Also, we will come to know the preferences of people, which will eventually help us to finalize that how we should do its branding. Accordingly, we can speak to advertize companies to prepare its various print and video advertisements before launch. Basically, we will test people as to what are the main qualities that they look for when it comes to superior skin care razor that they are not getting from currently available razors in the market. Since we are making different groups while doing stratification sampling, so the results can be different from its strata of people, which we need to collate and take similarities out of them.

Then form the data, research will be conducted on it to gather relevant information from the raw data as per the requirements. In this way, sampling will help us a lot to get relevant information required to launch a new product in our target population.

Bibliography

Boyd, N., 2015. What is Sampling in Research?. What is Sampling in Research? - Definition, Methods & Importance.
Buskirk, T., 2016. Target Population and Sampling Frame in Survey Sampling. Target Population and Sampling Frame in Survey Sampling, 21 January.
Gulf War Veterans, 2012. Definition of the sampling frame and sample, Available at: https://www.rand.org/content/dam/rand/pubs/monograph_reports/MR1018z12/MR1018.12.appb.pdf
Jewel, 2012. Sample designs and sample procedures. Sample designs and sample procedures, 14 July.
Kruse, J., 2009. Sample Frames. Sample Frames, Available at: https://www.iser.uaa.alaska.edu/Projects/living_conditions/sample_frames.htm
Polit & Hungler, 2015. POPULATIONS AND SAMPLING. POPULATIONS AND SAMPLING, Available at: https://www.umsl.edu/~lindquists/sample.html
Royal Geographical Society, 2016. Sampling techniques. Sampling techniques, Available at: https://www.rgs.org/OurWork/Schools/Fieldwork+and+local+learning/Fieldwork+techniques/Sampling+techniques.htm
StatPac Inc., 2014. Survey Sampling Methods. Survey Sampling Methods, Available at: https://www.statpac.com/surveys/sampling.htm
The Pennsylvania State University, 2016. Simple Random Sampling and Other Sampling Methods. Simple Random Sampling and Other Sampling Methods.
The Regents of the University of Michigan, 2010. Sampling Frameworks. Sampling Frameworks.
Turner, A. G., 2003. Sampling frames and master samples: UNITED NATIONS SECRETARIAT, Available at: https://unstats.un.org/unsd/demographic/meetings/egm/sampling_1203/docs/no_3.pdf