Bio-Statistics Research Activities, Whether Clinical Trial Based,

¶ … Bio-Statistics

Research activities, whether clinical trial based, experimentally designed, or product oriented, must exhibit and command interest, enthusiasm, and passionate commitment. To this end the researcher must catch the essential quality of the excitement of discovery that comes from research well done. The first step in the attainment of the desired research goal is to develop a scientific approach toward that which is being investigated. A requirement within the scientific approach best-fit format that is oftentimes misunderstood, and consequently wrongly applied, is that of sampling.

In a rather philosophical approach to sampling Ohlson (1998) states that sampling is " ... But part of the whole. Check to make sure I fairly represent my larger connection " (p. 27). With these words Ohlson is informing the research enthusiast that sampling alone can skew testing results, infuse uncontrollable error into statistical processes, and violate the empirical premise under which the research investigation is being conducted. The remainder of this report will attempt, by way of definition and example, to describe the manner in which sampling is conducted to achieve the necessary investigation results.

Definitions

With Random Assignment Using a Probability Sample. Whenever the word assignment is used within the constraints of a research investigation it denotes the representation of variables by way of participants. In the random assignment using a probability sample the research investigator has designed a study wherein the participants have been drawn or selected from a larger population and assigned to a treatment and/or control group by way of random application.

Without Random Assignment Using A Non-Probability Sample. In research investigations such as these the research investigator might possibly not have the availability of a general population as a results of study limitations, cost factors, limited resources, or simply because the study is so tremendously unique and confined that a population is not available from which to draw a representation sample that all identified subjects must participate in the study. In order to control some of the intervening error the research investigator will assign the non-random selected sample randomly to the treatment and/or control groups.

Without Random Assignment Using A Probability Sample. This particular investigative situation is generally reserved for cases wherein the research investigator has a large enough population from which to randomly draw a representative sample to participate in a study that is seeking differences, relationships and/or effects between specific variables wherein the treatment is more important, or the focal point, of research investigation rather that the participant. As such the participants are randomly assigned to the treatment procedure to reduce selection error or bias.

Without Random Assignment Using Non-Probability Sampling. This particular research design is reserved for investigative situations wherein a researcher is dealing with case study research or wherein only one independent variable is being investigated. The non-randomization of the sampling occurs when there is relatively no population from which to draw a sample or the investigator is embarking upon a new research frontier and is conducting what is generally called a pilot study.

Examples

Probability vs. Non-Probability Sampling. General speaking probability sampling is the selection of a sample from a greater population and is based upon the principle of selection by randomization of chance. This particular type of sampling is generally more complex, time-consuming, and costly. On the other hand non-probability sampling uses an arbitrary selection process and there exist no manner in which to estimate the probability of any one factor being included in the drawn sample. As such, there is no way to estimate sampling variability of identify possible bias. With probability sampling the research investigator, as opposed to non-probability sampling, is able to estimate sample reliability and sampling error and apply the research results to the overall population because the units of the sample are randomly selected.

Probability Sample Using Random Assignment: Available to the research investigator are several methods of probability sampling, namely, simple random, systematic, sampling with probability proportional to size, stratified, cluster, and multi-phase. As discussing all seven methods is beyond the scope of this assignment only one example will be presented befitting the simple random assignment technique. In simple random assignment sampling each member of the selected population has an equal chance of being included in the sample. In addition each member combination of the population has an equal chance of composing the research sample and it is these two properties that define simple random assignment. For example, consider a state's lottery. Most mega lotteries require generating 6 numbers form a population of 49 wherein each of the six numbers has an equal chance of being selected and each combination has the same chance of being the winning combination. Even though lottery players have a tendency to avoid combinations such as 22-23-24-25-26-27, it has the same chance of being the winning group of numbers as in a combination say of 7-13-28-33-37-46. The probability part of the equation would lie within the idea that 100 participants were randomly chosen from a population of 1,000 to determine whether or not any ticket holder held the winning ticket.

Non-Probability Sampling Using Random Assignment: As factors or elements are chosen arbitrarily in non-probability sampling there exists no way in which to estimate the probability of any one element or factor being included in the sample. In addition, the research investigator is offered no assurance that each item has an equal opportunity of being included in the sample and as such it is impossible to estimate sampling variability or to identify possible bias (i.e., difference between the true parameter value and the expected value of the estimator). The most common types of random assignment using non-probability sampling are convenience sampling (ad hoc), volunteer sampling, judgment sampling, and quota sampling wherein participants are randomly assigned to a particular group after being arbitrarily selected. An example of convenience sampling might be all female students (10) in the front row of a lecture hall and 5 are randomly assigned to participate in the treatment group and 5 are randomly selected to participate in the control group.

Non-Probability Sampling Using Non-Random Assignment: Oftentimes in experimental research "populations" are hypothetical and, as such, do not represent a true or potential population. In situations such as this the research investigator selects participants on the basis of what is available and assigns the individuals to certain groups using the random assignment method. For example, suppose a teacher wants to determine which teaching method, traditional or multiple intelligence is more effective with a class of 26 8th graders for reading improvement. In other words the sample becomes the population and the population becomes the sample -- non-probability selection as all students participate in the study. To achieve the random assignment requirement the teacher would randomly select 13 students to participate in each of the two teaching method's situations wherein all 26 participants have an equal chance of participating in either instructional method situation. The drawback here lies within the area of greater population inference making. That is to say, the results of the research investigation can only be applied to the 26 students and not to any other general populations such as all 8th graders in a school district, all 8th graders in a state, etc.

Probability Sampling Using Non-Random Assignment. Researchers primarily use this particular type of sampling when certain pre-selected independent variables are being measured. What is significant here is to randomly select the sample of participants from a greater population for the purpose of accurate representation and assign each participant to a particular testable sub-sample. For example, take for instance a research investigator who wants to determine, through comparison, the effects of a new anti-depressant drug vs. A placebo. The research randomly selects 100 from a greater population of individuals diagnosed as being depressed. The sample is then assigned to each group on the basis of the first 50 participants showing up for participation to the treatment (drug) group and the last 50 participants assigned the placebo group. Unfortunately in situations like this a great deal of investigative validity is lost.

Journal Article Research Critique -- Probability Sampling

Van de Berghe et al. (2001). Intensive insulin therapy in critically ill patients, New England Journal of Medicine, 345:1359-1367

Fairly critiquing an investigator's research endeavor is a task that must be taken seriously. Although it is quite easy to have an opinion of another's research it is something quite different to be able to evaluate the research activity in terms of topic specificity and soundness, intent or purpose, sampling, data analysis, and informational importance. The remainder of the first part of this review will be an analysis of a particular professional article wherein a group of critically ill patients were randomly assigned to two treatment groups (intensive insulin therapy and standard insulin therapy) to determine if normal blood glucose levels would improve the prognosis by use of insulin therapy.

The authors of the current project were, in opening statements, remiss in presenting a well-defined research question. Although the authors advised the reader that insulin therapy (standard and intensive) and adult mortality and morbidity was the focus of the investigation a meaningful presentation with respect to need was omitted. In other words, the authors did not build a medical or healthcare-based paradigm for the study. Following a well-defined research question the research investigators' task is to follow-up with a statement of a testable null hypothesis or hypotheses. The null form of the hypothesis is required in order for the proper application of a statistical data analysis tool to be implemented (Ohlson, 1998). Simply stating that an attempt to "confer" that an insulin deficiency might or might not affirm a "predisposition to complications" is not adhering to proper scientific research protocol with respect to a needed testable null hypothesis (Van Dalen, 1964).

Another important criteria of effective research are the use of an appropriate sampling procedure. Sampling alone can skew testing results, infuse uncontrollable error into statistical processes, and violate the empirical premise under which the medical process is being conducted. In order for accurate conclusions to be drawn about any medical procedure there must exist and adequate sample size as well as to employ the appropriate sampling procedure. In the arena Van de Berghe and his research colleagues were on target in randomly assigning patients to either the treatment or control situation. Although the research investigators chose to use the entire population as their sample this procedure can be described as a probability sampling as the sample is excessive (1548 participants). If the researchers had opted to take a random sample from the 1548 participants the procedure would likely have had delivered the same results -- provided the sample drawn from the overall population was large enough (Ferguson, 1966). Wherein they fell short was advising the reader as to possible unequal cell frequencies for the measurement variables they finally identified in the Outcome Section. Knowledge of the unequal cell frequency situation, and with respect to topical demographics, is important when the investigator chooses the appropriate statistical tool to analyze the measurement data. The research investigators might well have also informed the reader as to the type of randomization that was employed; i.e., simple random, stratified, cluster random, or equal probability random sampling.

Journal Article Research Critique -- Non-Probability Sampling

Thompson, David M., Kozak, Sharon E. And Sheps, Sam (1999). Insulin adjustment by a diabetes nurse educator improves glucose control in insulin-requiring diabetic patients: A randomized trial. CMAJ, 161(8):959-62