Friday, March 13, 2009


Today we spoke a bit about the fine art of sampling. Some of the terms are easily recognizable while others are a little harder to recognize and understand. Therefore, I am attaching the link for a website that does a very good job of simplifying the process of sampling in research. Click the link to check out the site.


  1. Sampling is important for research in that it saves time and money. From reading the information on the link, it does appear that it can be tricky in order to collect valid samples. There are concerns related to biased samples, measurement errors, the interviewer’s effect, the respondent’s effect, and knowing the study’s purpose. When conducting a study it is important to collect valid information so that the study is accurate. Researchers should take great precautions in determining if the information collected is good since their reputations are on the line. What good is a study if something is determined based on faulty information that is unrepresentative of the population or phenomenon being studied? Researchers must take careful consideration in which methods they used in order to collect samples. Before taking this research class, I never really thought much about sampling. I would have thought that the people you see in the malls doing interview sampling were taking a good, random sample. Now I realize that there is a lot more to it and that biased information can be reported even in situations such as this. For example, in the interview type sampling, a respondent may respond to a question based on the way the interviewer asks the question. I have learned that there are many ways in which sampling can be unreliable.

    Original by Kristi Maddox

  2. I am really glad that you put this link on the blog. I was not able to attend class last week and this link really helped me understand the sampling process. What at first glance seems to be a simple process can be really complicated if you want to do all you can to get a good sample. This link explained well the reason sampling is used instead of a census. It is also clear that some sampling procedures are better than others but they all may result in samples that are inaccurate and unreliable. The examples given were also very helpful. You are able to relate to what they are talking about and get a better picture as to why or why not a type of sampling is more or less reliable. Sampling error which comprises the differences between the sample and the population is due to the particular units that just happened to be selected. The example used about the average height of women was a great example to show how selecting the particular units is essential to the accuracy and reliability of the sample. I can see how the cost effectiveness of sampling is an important issue when determining how you are going to select your sample. If you make the wrong choice then all money invested is wasted.

  3. Original Post
    Malinda S.

    The sample size is one of the most important items in research. The sample size helps determine if the results can be generalized to the population from which it was sampled. Of course sampling can save money and time, and may be more accurate than actually involving every person or class in the research. Sampling is particularly important when sampling populations that are difficult to access. There are other items that are equally important, such as, the instrument used and reducing bias and or errors. Researchers usually take note to reduce the errors to increase the reliability and validity of their research.

  4. Sampling in Research
    Original Post: Tameka L.Askew

    The tutorial on Sampling was very interesting; it was full of informative information. I think that sampling is a very important factor in research. I gathered from the article that in sampling, research is not done on the whole population, but on parts of the population; sampling also only studies certain properties of a population. I found the information on sampling in research insightful and full of facts that I was not aware of; sampling also uses qualitative methods. In reading the tutorial, I gather that sampling is more practical and economical than a census. When I think of census, I think of annoying people knocking on my door. It appears that the census studies the population as a whole.

    This tutorial was an eye opener to the many different types of sampling. I would have to ask myself the question, which type of sampling is appropriate to use for a particular study? All and All, it appears that sampling research is economical and time conscious, I think that these are key factors to a project being successful.

  5. Malinda S. response to Diane Watson
    It appears that the sampling process can really be cumbersome. Just think of the implications of sampling gone awry if it was used instead of a census. What if the government and the IRB did not have adequate oversight over the process? Some areas would still be underrepresented and others overrepresented. That could cause the underrepresented population not to receive needed services. There is a possibility that the overrepresented population may start to feel really out of touch with the part of the country that needs the help.

  6. Malinda S. response to Tameka Askew
    Think of the money our government will save sampling instead of taking a census as in its present format. Think of the money that will be saved and could be spent providing some of the help that is so desperately needed. I would like to think that the money saved by sampling the population would be better served by bringing the schools into the current century.

  7. Original by Shija Brooks

    Sampling is the process of selecting units (e.g., people, organizations) from a population of interest so that by studying the sample we may fairly generalize our results back to the population from which they were chosen. Sampling information is obtainable quickly and less costly, but is not always the best route to take in research. It does not accurately represent any given population. Sampling can be used as long as the best sampling technique is used with the information being acquired.

  8. Original Post by: Julie Lindsey

    Sampling is absolutely necessary in research. When researching a topic involving a large population, it is almost impossible to interview every individual within that population. The example give in class by Dr. Vickerstaff involved conducting research on homeless individuals. There is no way that one would be able to locate every homeless individual due to inaccessibility.

    After reading this article, it was reiterated to me the fact that in order for research to be effective, research has to be a well thought out process. The researcher has to know exactly what they want to find out and how they are going to go about gathering the information. Researchers also have to proactive in anticipating any problems, such as sampling errors including any possible bias. Researches also have to take into consideration the sampling size. Research is a very big process that cannot be taken lightly.

    Julie Lindsey: Response to Melinda S.

    I was thinking about those populations that are inaccessible and how difficult it can be to reach them. Often times these inaccessible individuals do not want to accessible and if they are assessable, they may not be willing to participate in a study. Some populations that come to mind are homeless individuals and individuals living with AIDS. I can see how some studies involving inaccessible populations would be extremely difficult to conduct and would involve a lot of “outside of the box” thinking.

  9. Response to Diane Watson
    Tameka L. Askew

    Diane I agree with that the article was very informational and it opened my eyes to why sampling is use verses a census. I do think that being cost effective at this point in time in America is very important; waisful spending is not acceptable in this day and time. I agree with you that the right choice needs to be made the first time around so that is no money being wasted.

  10. Original Post by Carrie Wells

    Sampling is making headlines in the news lately as Obama plans on appointing a sampling specialist to conduct the 2010 census. The census serves many purposes: to redraw congressional districts, to determine the composition of the Electoral College, and to determine allocation of federal funding. Many politicians are opposed to using sampling as a way of conducting the census because they say sampling “doesn’t work.” Of course if done accurately, sampling can be very effective in providing valid and reliable data. It is the accuracy that sampling can provide that makes some politicians nervous. Republicans say door-to-door is the most reliable way, but democrats have advocated for the use of sampling so more minorities and immigrants will be counted. It’s interesting that the party with the least amount of minority constituents wants to use a method that provides the easier way to skew the numbers. Hmmm…is that gerrymandering I smell?

  11. Response to Malinda S. response to Tamika Askew
    By Carrie Wells

    I read that the cost of the 2010 census will be $15 billion…BILLION!! I just don’t understand how that can cost so much. We definitely need more transparency in our government. I wonder if it would cost less if sampling was used as the method of generating the census. The article I read this figure from was basing it on the paper
    mail-out system that has been in use for decades. Is it because of the rising costs of stamps? Maybe the government should invest in the forever stamp. ;)

  12. Patricia Ford
    Original Post

    Let me begin by saying I was so pleased with my ability to understand the article for the most part. I think we should all give ourselves a round of applause for learning to decipher some of this foreign language. Seriously, random sampling appears to be a very useful technique to utilize when conducting research. Sampling an entire population could be exhaustive and very costly. Random sampling allows the researcher to interview a sample of the population in an effort to make inferences regarding the population based on the information gathered. Doing this is not as cut and dry as it sounds. There are many factors that need to be considered. First of all, do you know your entire population and is it feasible to obtain a random sample. If so, what random size are you going to use? Ensuring the random size is large enough will reduce the chance error. However, you don’t want it too large either. Researchers must also take measures to minimize the risk of sampling bias and non sampling errors. A lot of time, effort, and consideration go into selecting the various types of samples. Not selecting a sample that is truly representative of the population may negate the researcher’s ability to generalize findings to the population.

  13. Patricia Ford
    Response to Carrie Wells

    Smells a little fishy to me. I appreciate your view on this issue. I have not had the opportunity to watch the news or any information on that topic. However, I would like to say it is pleasing to know the information we are learning is so valuable. It appears acquiring knowledge of this information will not only assist us within our profession but will also help us understand what is going on in the world around us. I admire the way you are willing to step outside the box and apply the information to the world around us. Kudos...

  14. Response to Carrie Wells by Kristi Maddox

    Carrie, you make some very interesting points in your post regarding Obama’s proposal. Taking a sample instead of completing a census would be a much more cost effective way to get results and we definitely need to seek to save money if at all possible. It would, however be difficult to obtain a sample that is representative of the general population. One would have to be very skilled, an expert, at sampling in order to obtain samples representative of the various racial and socioeconomic statuses of the population as a whole. This could prove difficult and tricky to say the least. However, if this was feasible to do correctly, I agree with you about doing the sampling instead of the census.

  15. Marashia McCormick
    Original Post

    Josh, I really appreciate how straightforward the information is on this site. One thing I’ve observed about the terminology in this course is that there are a lot of fancy words used to say some of the simplest things. You told us that research would be like learning a second language on the first day of class, and you were right. I just wish there were similar websites for COPES, MOLES, and design instruments.

  16. Marashia McCormick
    Response to Malinda S.
    Malinda, I don’t mean to be critical, but the census is not a population study it’s a sample one. We all know that they’re not getting a true depiction of the entire U.S. population. I personally believe they’re not even capturing a significant portion of the working poor, minorities, or homeless people. All of which are in desperate need of social and economic resources. I agree that a random sampling would be more effective, but there’s still no way of guarantying a 100% a response rate. The data will always be flawed.

  17. This article does a great job of breaking down the concept of sampling. This article also addresses the importance of paying close attention to sampling from the beginning. As seen through the examples, if sampling is overlooked there could be great consequences. Prior to studying research, I had not realized the importance and the depth of sampling. I think that it’s also important to realize that even when someone pays really close attention to his or her sampling techniques it could turn out incorrect. I think that part of the research process is being able to be intentional yet flexible. Set out to do the sampling right and if it doesn’t work out, be able to adjust and move on.

  18. With technology as it is today you would think that the census would be electronically run. A census may be more accurate, if people answer honestly. And if they don't is a census more representative than a sample? I may speak with ignorance here, as I don't know a lot about the benefits of a census. My question is "Is it worth $15 billion? It seems to me that this money could be used more efficiently, but I'm no government expert by any means...

  19. Original Post by Brent Eubanks:

    For the novice research student (myself), this article did an excellent job in explaining the process of obtaining a sample and explaining the types of samples. The article did a good job in attemtping to incorporate the reasons why the sample, in of itself, is so important to research. I also like the fact that the author used good examples of "research" that helped one undestand the process better. For example, the author, in discussing the TIME FACTOR, gave an example of a doctor that is having to gather information due to a disease breakout. I never really thought about the urgency of a situation like this and how even though you are not thinking about research and samples, per se, this is actually what is occurring-research.