Quality isfundamental in any type of research. The two aspects of quality in research arereliability and validity. There are types of measures used to assure that thestudy is reliable and valid. Sampling pertains to the determination of the whoand what should be included in the research.
These are the subjects who will beobserved, interviewed, or tested during the study. There are attributes that constitute trustworthinessor rigor of a research such as credibility, transferability, dependability, andconfirmability. There are two case study samples that are reviewed in terms ofits sampling, reliability, validity and trustworthiness. Sampling,Reliability, Validity and Trustworthiness Part of the planning phase ofresearch is to aim at sampling, reliability, validity, and trustworthiness inresearch. The benchmarks are established to assure reliability and validity ofa research study. These terms are further defined to understand theirdifferences and purposes as applied during research process. One important goalin research is to introduce randomness and prevent application of a system thatcould include biases.
Application of Sampling, Reliability andValidity on a Case Study Review Thereare two major types of sampling: qualitative sampling and quantitativesampling. There are sub-types under each type. Sampling consists of who andwhat to be included in the test. There are two general questions that must beaddressed prior to selecting your required samples: the size and thecomposition (Tappen, 2016). Sampling is also defined as the process of findinga representative of a chosen population for the intention of establishingparameters or attributes of the entire population (Robert, 2015). There arecertain terminologies that are constantly used in research sampling which mustbe identified and understood beforehand (Harvey, 2012-17). First is population. It is defined as the groupof people a researcher is interested in including in the study.
Second is parameter. It is defined as characteristicof a certain variable regarding the population. Third is sampling frame. It is the list of the entire population theresearcher is interested in. Fourth is sample.It is the subgroup of people from a sampling frame or population. Fifth is representative. It refers to the samplethat has the same attributes of the population.
Sixth is generalizable. It is the ability to generalize the data to thepopulation. Seventh is sampling error.It is the difference between a parameter and a statistic. Theprobability sampling and non-probability sampling are two types of quantitativeSampling. Non-probability sampling is like convenience sampling but non-generalizable.On the other hand, convenience sampling, purposeful sampling, and theoreticalsampling fall under qualitative sampling. Convenience sampling is the leastscientific and lacks intellectual credibility while theoretical sampling isbased on grounded theory which is more discriminant.
Theterm reliability can be replaced by numerous synonymous words but according toR. Tappen (2016), it involves research study or survey that produces consistentresults. There are questions a researcher must ask to address reliability. Forinstance, if two subjects in the same scenario use the same measure, will onearrive with the same results? Or, what if two different situations were used,would the results be the same? These are just couple of questions among many tomeasure reliability.
Validitypoints at how accurate a study or survey is at measuring what it is attemptingto measure. It emphasizes the measure being used is accurate to its deliberatepurpose (Tappen, 2016). The validity of a measure in research is deemedcritical. The goal of validity is to be as accurate as possible. There are severaltypes of validity: face validity, criterion validity, construct validity, andcontent validity.
Sample Case: Stress Management for Patients with Rheumatoid Arthritis (Prime,n.d.) Thiscase study is about a 40-year-old woman who has been followed up by a nursepractitioner for rheumatoid arthritis (RA). She was diagnosed a year and a halfago for moderate RA and was prescribed with an antirheumatic drug. About twomonths before her visit to the clinic, she experienced flare-ups of hersymptoms (pain and swelling).
At that time also, she was experiencing morestress at work after receiving a promotion. Her diagnostic tests includedincreased inflammatory hallmarks. She verbalized her worries of finances andshe was curious whether these symptoms are attributed to her current stresslevel. She also wanted to know if there are other avenues to lower her stressand reduce her symptoms. Thebackground research done on this study revealed that RA affects a smallpercentage of the population. The presentation of this autoimmune diseaseincludes damage on the joints which can eventually lead to considerabledisability and with a possibility of mortality. Clinical evidence shows thatadded psychological stress can increase these symptoms. There is a strongconnection between minor stresses and exacerbation of RA.
A 5-year study alsoproved that those patients with higher stress level experience more jointswelling than RA patients with lesser stress. On the hand, another study showsthat RA patients experiencing disturbances in mood have higher inflammatoryattributes with joint pain. Some research shows that behavioral interventions canprovide safe and money-saving benefit to RA patients. Some of these behavioralinterventions include cognitive behavioral therapy (CBT), tai chi, yoga, andstress management education.
Discussion: Based on the informationavailable in this case study, the sampling technique used is qualitative samplingwhich could either be purposeful sampling or theoretical sampling. Purposeful sample is the sample ofparticipants, those with RA condition, who are purposefully considered for thestudy because of their common experience with RA. On the other hand, it couldalso be a theoretical sample, where the sample of participants, because oftheir diagnosis-related experience can help in the development of a theory (forexample, RA patients who perform tai chi presented an increase in endurance andstrength of their lower limbs). Thereare discrepancies and inconsistencies in the results of use of behavioral interventionssuch as CBT, tai chi, yoga, and education.
Some are inconclusive and needfurther studies due to the limitations of the study. Other results suggest thatalthough the benefits are unquestionable, the long-term effect and degree ofeffect of these interventions are yet to be studied further. Asevident by the inconsistencies of results, it suggests that the reliability andvalidity of this case study is a concern.
In terms of reliability, there arenot enough details regarding the process on how the behavioral interventionswere conducted (for example, different amount of time applied on the study ofyoga and CBT). Also, not enough information on how samples (patients with RA) aretested or rated. In terms of validity, is the study accurate enough inmeasuring what it is trying to measure? Unfortunately, in this study, there aremany factors such as pain ratings, levels of anxiety, anatomical differences,degree of RA symptomatology, different kinds of stressors, should have beenconsidered before the study. Rigor in Qualitative Research(Trustworthiness) Qualitymust be ascertained in any type of research. Quality is fundamental. Rigor is referredto the compliance to the quality standards in conducting research to increasethe assurance in the results. To assure rigor in research, the method inanswering the intended research question must be appropriately addressed andthe chosen research design must be strong.
Researchers believe that withoutproper rigor, a research is bound to be no worth. Therefore, there must be ageneral criteria or quality measured standards in evaluating research studies. Rigorin research is attributed to the trustworthiness of the research.Trustworthiness in qualitative research describes four basic elements:credibility, transferability, dependability, and confirmability (Tappen, 2016). Asdescribed by R. Tappen (2016), credibility is identical to internal validitywhich can be accomplished through five different ways.
First, is by having enoughtime to persistently engage, observe, and test for possible reason and outcome.Second, is by reviewing and sharing your raw notation, interview transcripts orpreliminary outcomes to your participants or representatives of the people you areworking with to gain input or feedback from them and later include thosefeedbacks in your conclusion. Third, is by including feedback from experts inthat particular study: methodology or subject. An objective critiquing of astudy takes time and interest on the part of the expert but can yield atrustworthy result.
Fourth, is by providing and considering other possible explanationson participants’ negative viewpoints or data. The existence of the negativecases must not be ignored and must be analyzed and reported regardless of itsimpact. Fifth, is by referencing or cross checking two or more data to provide widerperspectives that can be valuable in your complex study.
Transferabilityas noted by R. Tappen (2016), is parallel to external validity. In quantitativeresearch, transferability refers to how findings are applicable to other peopleand situations thus can be generalized. Dependability, which is similar to reliability,can be achieved by creating an audit trail while research is underway. Audittrail covers a recorded compilation of your notes, researcher’s thought processand reflections, survey guides, data, and including decisions during the study.Confirmability as R. Tappen (2016) describes is the objectivity specifically inquantitative research.
The inclusion of a reflexive journal, which includes recordedexpectations and ideas prior to the study, which can support a researcher asreference while answering his research questions. Sample Case: Young Adult E-Cigarette Users’ Reasons for Liking and NotLiking E-Cigarettes (Pokhrel, Herzog, Muranaka, & Fagan, 2015) The objective ofthis study is to have a wider understanding of what young adults, ages 18-35, whoare users of electronic-cigarettes, like or dislike about electronic-cigarettesor e-cigarettes. This study focused on determining the reasons of encouragementand discouragement from the use of e-cigarettes. There were focus groupdiscussions conducted. Data gathered and analyzed with results indicating 12 classificationsof rationale on liking e-cigarettes while 6 classifications of rationale on thedislike of e-cigarettes. The study’s conclusion provided the motives of theyoung adults for the use and not use of the device.
Their motives vary inrelation to the importance of its use. The findings also show that e-cigarettesusers, although purposefully served for social and recreational, do not dependon just the nicotine effect. In addition, this study also discussed theefficiency of the delivery of nicotine, smoking cessation, and devicemanagement. Discussion: This study provedto be trustworthy and here are the reasons why. In terms of its credibility, a 4-month study wassustained sufficient to engage and observe the participants. Triangulation wasalso used in this study, allowing data cross-checking for inconsistencies.
Theresearchers of this study, not only included the strengths of this study butalso included the limitations which constitute negative case analysis. In termsof dependability, the investigators,coders, and note takers have used questionnaires, surveys, notes,audio-recorders, and have reviewed their transcriptions and notes. They alsodiscussed areas of agreement and discrepancies. In terms of confirmability, just like the audittrail, the researchers used a reflexive journal where methods and datacollection were logged including nonverbal responses of the participants.
Although personal thoughts or preliminary reflections by the researchers werenot stated in the study, reviewing of their transcriptions regardingdiscrepancies and addressing them was mentioned in the study. In terms of transferability, this qualitative researchcannot be generalized. The results are specific to the focus age of 15-35, thuscannot be validated externally. Nonetheless, the overall trustworthiness of thisstudy is high. Conclusion The type of sampling utilized inresearch affects the accuracy of the results. A qualitative sampling was usedin the first sample case study. Although, in this same study, the validity andreliability are in question as evident by the inconsistencies of results. Inthe second sample case, the trustworthiness of the study is proven high basedon the parameters of quality rigor in research.
Validity, reliability, andtrustworthiness in research are essential.