A research sample is considered not necessary when dealing with a population that is considered to be under the same authority. That is a population that follows a directive from the same chain of command. Such a population it is believed would follow the same route and give the same answers to an inquiry made as to their mode of conduct. This kind of assumption was made, for instance, when researching about the conduct of the so-called disciplined forces. Its results were practical, for example on Stanley Milgram's research on obedience to authority. To prove to the readers that the Germans were in particular obedient to the authority. That was a common viewpoint for the Nazi killings during the Second World War. He in particular wanted to know how someone would act when ordered to do something by a higher authority. His case was made simpler by the defenses of those accused of murder during that period. The respondents in the case summarized their cases that they were simply acting out of obedience to the authorities. The outcome of his research was believable and stressed the assumption that population acting under authority can be presumed to have a common reason to justify their actions.
Alcohol reaction to the human body would also vary from one person to the other. Depending on the factors and the circumstances one is taking the alcohol, reactions of their bodies are bound to differ. While researching on this topic, it would be advisable that a bigger representative sample was taken so as accurately to represent the different reactions of their bodies.
It was possible though to make assumptions on the likelihood patterns of voting among a population. Where the elections are issue-oriented, the issues would be the same across the board. It would be difficult to isolate the issues from the population. In this case, the issues would be like an authority directing the population to vote in a particular pattern. In this case, therefore, it would be right to assume that the population is homogeneous.The reports of the sampled population indicated the total number of persons interviewed. It also states how that number was arrived at and its composition by age in percentages. Their resident states are also quoted to verify that the research covered the whole country. According to McCarthy (2014), a percentage of the margin error is also stated to add substance as to the authenticity of the results and the survey. The means of reaching the respondents were also stated. That was to prove that there was no partiality in the identification of the respondents. Daily News adds that ‘additional minimum quotas by time zone within a region.' McCarthy (2014), USA being a cosmopolitan country the sample selected ought to have stated the composition of their sample as either immigrant or Native Americans. Gender composition too ought to have been considered to satisfy the population that the sample was ideally representative (Chambliss & Schutt, 2010).
There are different reasons why a researcher would choose a sample population. The method of choosing a sample population also depended on the reason for carrying out the research. The article opted for simple random sampling. In their case, the research population was known and was available to the researchers. The method was also necessitated by the need to carry out statistical analysis after the research. Based on the sampling methodology the researchers argue that random sampling is not so prone to errors and would give almost accurate results. That is they would give the smallest margin error of the desired outcome.Probability-based sampling designs are highly representative if all the respondents were interviewed. It correspondingly ensures proportional representation by, age, gender, and class. Contrary, a conclusive research using this sampling technique without a complete list of the sample population. Isola6ting members of the same group so as to interview them can also be viewed as divisive. Probability-based sampling can also be said to be time consuming and also complex when stratified random technique was applied.
Probability-based design could not have achieved the desired results as was the non-probability based design. That was justifiable because the research populations was known and defined. Assuming that the population was to be picked randomly, the desired results could have had a larger margin error. Using a non-probability based design in sampling does not produce near accurate results. Errors were numerous the results to be justifiable. Without a doubt, it was cumbersome for a sample population to representative of the research population.
Using a research topic on the adaptation of first years to college life, sampling components refers to the parameters that define a sample population. For example, they would include gender age, social status, level of education and even social background. The diversity would include social background (Chambliss & Schutt, 2010). That was because, and not all the students came from one part of the country. Even though they all qualified to join the college, not all of them underwent the same challenges in life.
In choosing a research sample, I would use probability sampling because the research population was known and well defined. In researching about the church introducing a new type of music, the research population would be the 150 worshippers. The population parameter would refer to the average number that if sampled would give the representative view and in this case would be 15. Sampling unit would be a single unit of the whole research population and in this case would be 1. Sample statistics would be the information derived from the sample population. In this case, it would be the wishes of the 15 worshippers chosen from the church.The research articles for research project's researchers decided on their samples based on the information they intended to achieve.
Psychology research is basically a pre-informed position that a researcher or a psychologist wishes to put to the public. In this way, non-probability samples would allow him/her to manipulate the results to his/ her favor taking into consideration their earlier taken positions.
Errors in probability-based sampling increases as a result of undefined research population. In stratified, proportions of subsets must be known for accuracy. Disproportionate was more biased. Stratified sampling is very flexible and can be applied in many geographical surveys. Disproportionate is more biased and therefore favorable for research topics whose outcomes had already been pre-determined. Simple random can lead to under-representation of the target research population.
Probability-based sampling techniques are largely representative as opposed to non-probability based sampling. Complexities in its management are very minimal. They are also equally representative as opposed to non-probability techniques where it is hard to tell whether the sample is representative of the research population.
References
Chambliss, D. F., & Schutt, R. K. (2010). Making sense of the social world: Methods of investigation. Los Angeles: Pine Forge Press.
McCarthy, Justin (2014). Most Americans Still See Crime Up Over Last Year Retrieved on 22nd November 2014 from: http://www.gallup.com/poll/179546/americans-crime-last-year.aspx
