tendency for novice researchers to develop their own instrumen

There is a tendency for novice researchers to develop their own instrument if they cannot readily find one. How might you respond to a peer or manager who asks you to help develop a new tool to collect patient data on anxiety prior to cardiac catheterizat

LECTURE NOTES
Analysis of Qualitative and Quantitative Data
Introduction
Quantitative studies utilize structured data for measurement of all of the variables of interest in a study. Structured data is desirable because it allows for consistency of measurement, enhancing the objectivity of the data and reducing bias. The quality of the measurements used for data collection is foundational to the value of research findings. The methods used to analyze data differ significantly between qualitative and quantitative data. Qualitative data analysis can be challenging for the researcher because there are no set rules in place which can make it difficult to explain how one should do the analysis. This lecture will take a look at both approaches of data analysis and the collection and measurement of data in quantitative research.
Quantitative Measurement
Measurement, in simplest terms is assigning numbers to quantify the presence of an attribute. Measuring weight and temperature are familiar examples that utilize rules for measuring (Polit & Beck, 2012). The strength of a measurement in quantitative data is that it completely removes subjectivity and guesswork. This means that the information obtained tends to be objective and it can be independently verified.
Strategies for Collecting Structured Data
According to Polit and Beck (2012), “structured data collection involves having a fixed, rather than flexible, approach to gathering information. Both the people collecting the data and the people providing the information are constrained during the collection of structured data” (p. 191). This is in sharp contrast to qualitative studies where the researcher’s goal is to learn from the perspective of the individuals who are living the experience. The use of structured self-report instruments is the main strategy used in nursing research.
The Use of Questions
The simplest questions used to collect data are closed and open-ended questions. There are many ways to format closed-ended questions. They can be dichotomous (e.g., yes or no) or multiple choice. Subjects can be asked to rate the response that is of greatest importance or asked to list the responses in order of importance. A visual analogue scale is another example of closed-ended questions: the rating of pain from 1 to 10 is a visual analogue scale. The most well-known scale is the Likert scale, where respondents are asked the extent to which they agree or disagree with a statement.
Observation
Data can also be collected via observation. There are some variables that can only be measured through observation. If the focus of a study is infant behavior in response to a specific intervention, observation may be the only way to collect data. In order to prevent bias, the development of guidelines, checklists, rating scales, and training of observers is essential. Bias is a particular threat when observation is the means of collecting data.
Biophysiological Measurements
The advantages of these types of measurements are the accuracy and validity of these measurements. However, beyond the use of basic instruments such as scales, thermometers, and stethoscopes, the use of technology may be cost prohibitive and beyond the scope of nursing practice to obtain.
Assuring Quality of Measurement and Data Analysis
It is helpful to understand that all measurements have some error component. For example, test scores may or may not be a valid measurement of knowledge. Tests always include some but not all of the information to be learned. The test items may lack clarity or specificity; wrong answers may be chosen inadvertently or accidentally. Personal factors such as fatigue, anxiety, and other distractions may also cause selection of the wrong answer even though the information being tested has been learned. These same types of factors affect measurement in research studies. The goal is to measure the truth and minimize error.
Data analysis of a quantitative study needs to assess whether there is similarity between the research variables as discussed in the introduction of the report and as described in the method section. The report should include the reliability and validity of the measures and that the evidence came from the research sample and not from other similar studies. If the data is extremely flawed, the study cannot be considered useful evidence. This may create questions to research consumers that the study does not provide accurate data and that the key constructs would not seem valid (Polit & Beck 2012).
Analysis and Interpretation of Qualitative Data
The analysis and interpretation of data requires refined skills of the researcher. The process of analysis and interpretation occurs simultaneously, as well as after the conclusion of data gathering and analysis in order to formulate appropriate conclusions. Although the steps occur to some extent simultaneously, they will be described separately.
Analysis of Data
The analysis of qualitative data requires large amounts of data to be reduced to a small number of themes or concepts that represent the phenomenon for the participants. The majority of data is taped or collected through interviews. The tapes are transcribed verbatim, capturing pauses and vocal noises that may indicate confusion, surprise, laughter, and other emotions. If someone is hired to transcribe the data, instructions must be very clear that the transcription must be identical to the spoken words, as well as other sounds, pauses, and interruptions. In other words, the transcription should reflect the interview to the fullest extent possible. This includes the careful use of punctuation so as not to distort the original meaning. During the interview, the interviewer may jot down notes where observations may become critical to interpretation of the text. The researcher must be immersed in the data, reading hundreds or thousands of pages of material and/or listening to the tapes multiple times (Polit & Beck, 2012).
As themes emerge from the data, the researcher writes them down in order to form categories. The analysis of data may be started after the first few interviews. As themes begin to emerge, interview questions may be altered in order to assure that relevant information is collected in later interviews. If information is uncovered in later interviews, typically, the researcher will return to the earlier participants to have them confirm the new information.
When data is managed directly by the researcher, several methods may be used to reduce the data. This reduction of data requires the researcher to thoughtfully consider assumptions and implications. This will require reflection, posing questions relating to the data to test developing assumptions and relationships, consulting with participants, conversing with peers and mentors, and returning to the data numerous times. The process of data analysis could last from weeks to months. Once themes have been uncovered, they may be color-coded in order to find them more easily before they are transferred to computer files or index cards. Once these initial categories have been identified, they may be further reduced to more abstract concepts. If relationships exist between concepts, they will contribute to the development of theoretical statements. If possible, the data may be reduced to one overarching construct that represents the phenomenon of interest.
Interpretation of Data
Immersion in the data is required for interpretation. In addition to reflection on the data, the researcher also must reflect on personal beliefs, values, and worldview to limit bias in reaching conclusions. According to Polit and Beck (2012) creativity is extremely important to understanding the meaning of the data. The researcher needs to give himself or herself sufficient time to achieve the “aha” that comes with making meaning beyond the facts. Communicating further with participants, searching the literature for relevant theory and research, consulting with individuals with specialized knowledge, and peer review all help to answer questions that arise about meaning and relationship.
Enhancing Integrity of Qualitative Research
There are several strategies to enhance the trustworthiness of results. The first of these is bracketing, referring to the researcher’s efforts to identify personal values and beliefs so these can be held in abeyance when collecting, analyzing, and interpreting data. Prolonged engagement assures that the researcher will capture the fullness of the phenomenon and prevent premature conclusions. Triangulation, where additional sources of data or information are included, allows for confirmation of the findings. Member checks, or informant feedback, allow participants to provide comments as to the accuracy of the data or interpretation. Conducting an audit trail, including keeping a diary of processes, thoughts, and analysis of decisions made, allows peers or others to evaluate the extent to which the researcher implemented required strategies to assure credibility of findings.
Transferability of the findings means that the findings of the study can be applied to other populations. This is similar to external validity in quantitative research. Transferability depends to a great extent on the sample. Sample characteristics (e.g., gender, age, education, culture, race, geographic location, socioeconomic status) are used to identify similarities and important differences.
Conclusion
Data analysis and interpretation requires knowledge and skills. Analysis of qualitative and quantitative data is accomplished according to rigorous criteria and rules. If these criteria are not applied, there is questionable credibility of findings. Quantitative studies require structured data. The mathematical level of measurement will determine the applicable statistical tests. Both qualitative and descriptive data describe aspects of the phenomena of interest. Qualitative data analysis results in the reduction of data into themes and concepts that capture the essence of the data described by words. Both types of research need a concrete study method to make the data meaningful.
References
Polit, D. F., & Beck, C. T. (2012). Nursing research: Generating and assessing evidence for nursing practice (9th ed.). Philadelphia, PA: Lippincott Williams & Wilkins.