Abstract
There are different research designs that we have. Different people prefer to use the different designs for different reasons. There are threats to validity for each of the designs. There are factors that can affect the research design that an individual or a group of people carrying out the research will prefer to use:
A researcher must first find out the best research design for the particular type of research they are undertaking. Different research works work different with the research designs. The individual will have to look at the available designs and see the advantages and disadvantages of each before choosing a particular design to use. Some designs may also be easier to use for specific research works. Cost remains as a fundamental factor to consider before settling on a specific research design. An individual or a group of people undertaking the research has a budget that they have to work with in the project. They will have to analyze the costs of using the available research designs. They will then have to settle on the design that fits their budget, and that is also very adequate. Time is an important for any research. Different researches require different time limits. There are those researchers that will take several years while others are just a matter of days. The time spent is dependent the scope of research. The scope is the area that the research will be undertaken. A larger scope will mean more sampling methods and more assessment objects, therefore using more time. When deciding on the design method to use, the researcher has to decide which research design best works for the type of research they are undertaking. The different research methods are suitable for different research topics as some take more time than others.
Study Design
The research design is a planned procedure to study a particular problem. It uses different things such as formulating research questions, a hypothesis analysis, the critical analyzing and formulation of independent and dependent variables, the experimental design and the methods of data collection and a statistical analysis plan (Robson, 2011).
Research designs available include; longitudinal design, , the experimental design, the case study design, the action research design, meta-analysis design, cohort design, casual design, cross-sectional design, descriptive design, exploratory design, historical design, observational design, philosophical design and the sequential design.
For validity, we have both internal and external validity. Internal validity is specific to whether an experimental condition makes a difference or whether it doesn’t. It also checks to confirm that there is sufficient or enough evidence to support the given claim. External validity simply to how possible it is to generalize the condition outcomes.
Our case uses the action design. It is where learning curve takes the form of a cycle. There is an understanding of the problem, which are the effects of tobacco on the human body. An intervention plan is made. The intervention is then carried out. During this time, important observations are made. This intervention is applied time and again, in a cyclic form until an implementation solution is found (Robson, 2011).
Advantages of action research design
The action research design is laid its focus on practical and solution-driven research. The research is collaborative and adaptive. It is mostly useful in community and work situations. The practitioners learn from the experience of what happens each time they apply a particular method. It is difficult to conduct the research since the practitioner researchers on the topic and also takes a responsibility in advocating change.
Disadvantages of this design
Here, the experience got which in most times takes the form of observation and stories is the data. A researcher can, therefore, give false information to fit their personal interests. The researcher may end up being over involved with the research that may cause bias in the results. The action research design is difficult to conduct and time-consuming. The researcher needs to make several observations that are being done in cycles before coming up with an exact conclusion (Robson, 2011). The participants also need to be cooperative to get the change that is being advocated, which is difficult to attain.
Study Sampling Strategy
When conducting research, it becomes difficult to sample all the target population. Researchers, therefore, get a group of individuals or objects from the target population to represent the entire population. Sampling all the target population is difficult as it is expensive and time-consuming. It can also cause inconveniences to the individuals under the study.
The various sampling strategies which are classified into two: the probability sampling technique and the non-probability sampling technique (Robson, 2011). The probability sampling techniques include the stratified random sampling, systematic sampling, simple random sample, and cluster sampling. The non-probability sampling techniques include the self-selection sampling, quota sampling, purposive sampling, convenience sampling and snowball sampling.
When choosing samples, the researcher should be aware of voluntary response samples, convenience samples and bias. Individuals, who most times volunteer for sampling, have a personal interest which will alter the sampling results. They most times give their opinions based on personal reasons rather than the research’s importance. Also, getting convenience samples gives bias as the most time give positive or the exactly expected responses (Robson, 2011). Approaching non-convenient individuals gives a researcher the information they need which will lead to making a better conclusion. Our article does not offer the sampling technique used. We can’t tell how many addicts were involved in the research. It is not also possible to tell whether those who managed to quit smoking had been long time smokers or addicts. They may have been young smokers who had only smoked for a few years, months or days. That is the reason it might have been easier for them to quit smoking compared to their counterparts who had been smoking for many years.
Threats to Validity
New Research Design
The cross-sectional design can be used. The design does not have a time dimension, and it is dependent on existing differences but not the change observed after intervention and groups are not indiscriminately allocated, nonetheless, the selection is based on existing differences. The design measures the differences from a variety of subjects but not the process of change.
Advantages of the cross-sectional design
At a particular stage, the cross-sectional design provides a snapshot of the results and the characteristics associated with the results. Since the sample is taken from the whole population, it is possible to estimate the outcome's prevalence (Robson, 2011). The sampling method used, which is based on existing differences and not collecting of a random sample provides a good outcome as the individuals are so different and only sharing one or two characteristics. The sectional studies also have the capability to use data from a large number of subjects as it is not restricted to a geographical area.
Disadvantages of the cross-sectional design
In the design, there is no follow-up on the findings. Apparently, it is difficult to use studies to establish the causes and effects of the relationships.
Threats
It is difficult to find people with very similar characteristics apart from one specific variable. The results are not in sequential form and do not reveal the history. It is because they are time bound. With this design, it is possible that if a different time-frame had been chosen, there is a probability that we could obtain different results.
New Sampling Strategy
We can find the tobacco users and the smokers, organize them in groups. The grouping can be based on the level of addiction or the time that the individuals have been using the tobacco and smoking. We can then monitor the change and make observations on what happens. This way, it will be possible to give feedback for the best control to give to smokers at certain levels.
Conclusion
There are many different research designs and sampling methods. Different research topics can use different designs and techniques. Some research topics may have more than one designs and techniques to choose from (Schreiber & Self, 2011). The researcher can decide the design and sampling methods they would like to use by weighing the threats, disadvantages and the advantages for using these designs and methods.
References
Robson, C. (2011). Real world research, 3rd ed. West Sussex, UK: John Wiley & Sons.
Schreiber, J., & Self, K. (2011). Educational research: the interrelationship of questions, sampling, design, and analysis. Hoboken, NJ: Wiley.