Introduction

If you have ever taken a survey after participating in some kind of experience or purchasing a product or a service, you have participated in quantitative research; the same is true if you have completed a questionnaire—even if said questionnaire was just a google form—or even have simply used the internet. If you have engaged in any of these activities, you have provided some amount of quantitative data to the world.

The National Library of Medicine provides an excellent overview of quantitative data: these data are “represented numerically, including anything that can be counted, measured, or given a numerical value”.[1] Quantitative data varies in nature, from categorical data (e.g., countries) to discrete data (e.g., number of students in a class or a school) to continuous data (e.g., a value within a range such as temperature or height). Data scientists collect quantitative data in a variety of ways including survey data, questionnaires, tests, and other similar methods.[2] Once collected, data scientists may employ any number of methods to analyze the data; most of these methods involve statistical analysis.

Quantitative data’s range is not limited to its status as categorical, discrete, or continuous nor how data scientists collect and analyze it. That is, quantitative data also spans a variety of topics ranging from behavioral science to health sciences. This discussion will center on quantitative medical health research, research which was likely collected through questionnaires in a doctor’s office or through similar circumstances. Read on to learn our recommended best practices for quantitative data collection in medical health research.

Best Practices for Quantitative Data Methodology

Perhaps the best place to locate best practice recommendations for medical health research is the Centers for Disease Control and Prevention. Known as the CDC, this key government health organization breaks effective quantitative data collection into two main categories: planning quantitative data analysis and conducting quantitative data analysis.[3]

Planning Quantitative Data Analysis 

Making a plan for your data collection ahead of time may make the process not only be likely to run more smoothly but also generate data that will be worth the effort.[4] Best practices for a quantitative data analysis plan consist of: identifying your purpose, identifying your analyst(s), identifying (or developing) your system for data management, and identifying (and/or obtaining) your data analysis software. 

Identify your purpose 

This part of the planning process is perhaps the most important. Identifying the purpose of a data analysis consists of determining whether and how each piece of collected data will facilitate improvement and understanding of a given field once a data scientist analyzes it. The CDC succinctly summarizes this concept, stating that analysts must “decide how [they] will use [their] data to improve your program before [they] collect them”.[5] To borrow a popular phrase from the business world, you must find the “why”[6] [7] for your data collection.

Identify your analyst(s)

 Also important is choosing trained and capable data scientists to analyze the data that you plan to collect.[8] Perhaps this seems obvious but many data scientists have particular strengths. Beyond quantitative, qualitative, and mixed methods scholars, many people in the analysis field have methods that they are best at or even just prefer to use. So, prior to collecting your data and definitely prior to choosing the data analyst you wish to recruit, ask yourself what kind of data analysis you wish to conduct once you have your desired data. 

Identify (or develop) your system for data management

The CDC also recommends that researchers develop a data management system before the data collection process begins.[9] Some popular choices for data management are spreadsheet software programs such as Microsoft Excel and Google Sheets. Whatever your data management system ends up being, the purpose of such a system is make it easier to maintain data quality during the data entry process and during the general management of the data once it is processed. Once the data has been entered into your data management software, you can commonly export files from this software directly into analysis software such as IBM’s SPSS program.[10]

Identify (and/or obtain) your data analysis software 

SPSS is one of many data analysis software programs. Besides SPSS, other advanced statistical analysis programs include SAS,[11] Stata,[12] and R;[13] some of these programs require purchase (but may be available through institutional access) but others—such as R—are free to the public. For simpler analysis needs, analysts may choose to stick with their data management software (e.g., Excel) to conduct these calculations.[14]

Conducting Quantitative Data Analysis

Once you have completed the necessary preparation steps and collected your data, you will then be ready to begin conducting data analysis. These four steps consist of identifying errors in (cleaning) your data, running statistical analysis tests, interpret and review your data, and summarize your data findings.

Identify errors in your data

Prior to conducting any sort of analysis, assessing the data for any sort of errors needs to happen so that final results are as accurate as possible; ideally, the data cleaning process is completed by multiple analysts to increase the likelihood of catching all errors.[15] This includes addressing missingness, duplicate values, extreme outliers, nonconsenting participants, invalid entries, etc.

Run statistical analysis tests

Once data are sufficiently cleaned, data analysts can then perform the necessary statistical analysis tests. Many times, basic descriptive statistics—used to assess the primary characteristics of the data—will provide the needed analysis.[16] Common descriptive statistics include frequencies, percentages, ratios, and measures of central tendency (mean, median, and mode). These analyses are particularly useful in learning about the demographics of the study participants (e.g., how many participants are teachers, mean age of the teachers, ratio of teachers to students, etc.). Of course, analysts may wish to perform more advantage statistical procedures such as linear regressions or chi-square analysis; the analysis procedures are determined when the data analysts determine the nature of the research.

Review and Interpret Results 

After performing statistical analyses, it will then be time to review and interpret the results of your data analysis. For example, analysts will assess for patterns within the data by considering differences and similarities between participants.[17] Which specific method of results interpretation will, of course, depend on which analysis method was used. While interpreting descriptive statistics (e.g., frequencies, measures of central tendency, etc.) is likely a more straightforward process, other analyses will require more advanced technique (e.g., interpreting the results of a linear regression or a binomial regression).

Summarize your data findings

Once you have completed the process of reviewing and interpreting the results of your analysis, you will then summarize your data findings; how you summarize your results will differ depending on what your ultimate goal is. If you wish to present your findings at a meeting or conference, for example, you will likely want to design visuals such as table, charts, and graphs to summarize what you found. Visuals are especially helpful in the health sector in that they help visualize the statistical analyses[18] and make them more accessible for the lay person.

Conclusion

Quantitative data collection—and doing so in a way that maximizes quality—is vital for medical health research. In other words, collecting high quality quantitative data provides a key cornerstone to the necessary foundation for addressing health issues. Best practices ensure the data collected is reliable and insightful, involving meticulous planning, skilled analysts, effective data management, and appropriate analysis software. The process continues with cleaning data, running statistical tests, interpreting results, and summarizing findings.

One critical application of quantitative data collection is in addressing health and healthcare disparities^[19]. Organizations with infrastructures for quality measurement and improvement, such as hospitals, community health centers, physician practices, health plans, and government agencies, play key roles by incorporating race, ethnicity, and language data into their data collection and reporting efforts. These efforts are essential for identifying and addressing disparities in healthcare access and outcomes.

By examining current practices and addressing identified issues, healthcare organizations can enhance their data collection and reporting. This enables more accurate and comprehensive analysis of health disparities.

Successful quantitative data collection in medical health research demands a comprehensive strategy that integrates detailed planning with rigorous analytical techniques. By incorporating race, ethnicity, and language data, and addressing the specific challenges of collecting this information, researchers and healthcare organizations can work towards ensuring that their data is more inclusive and representative. Following these best practices enables the medical community to draw meaningful conclusions that influence public health policies, clinical practices, and further scientific inquiry, ultimately improving health outcomes and advancing medical research.

Take Away

Effective quantitative data collection is crucial for reliable medical health research. Best practices include thorough planning, skilled analysis, and appropriate software use. Data cleaning, statistical analysis, and careful interpretation enhance result accuracy. Integrating race, ethnicity, and language data promotes inclusivity and helps address healthcare disparities. These strategies support informed decisions, public health policies, and improved medical outcomes.

[1] Network of the National Library of Medicine. Quantitative Data. National Library of Medicine. https://www.nnlm.gov/guides/data-glossary/quantitative-data

[2] Sadan, Vathsala. Data Collection Methods in Quantitative Research. Indian Journal of Nursing Education.

[3] Centers for Disease Control and Prevention. Analyzing Quantitative Data for Evaluation. Evaluation Briefs. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.cdc.gov/healthyyouth/evaluation/pdf/brief20.pdf

[4] Centers for Disease Control and Prevention. Analyzing Quantitative Data for Evaluation. Evaluation Briefs. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.cdc.gov/healthyyouth/evaluation/pdf/brief20.pdf

[5] Centers for Disease Control and Prevention. Analyzing Quantitative Data for Evaluation. Evaluation Briefs. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.cdc.gov/healthyyouth/evaluation/pdf/brief20.pdf

[6] Sinek, S. Find your WHY. https://simonsinek.com/books/find-your-why/

[7] Burns, S. What ‘Finding Your Why’ Really Means. Forbes. https://www.forbes.com/sites/stephanieburns/2021/05/24/what-finding-your-why-really-means/

[8] Centers for Disease Control and Prevention. Analyzing Quantitative Data for Evaluation. Evaluation Briefs. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.cdc.gov/healthyyouth/evaluation/pdf/brief20.pdf

[9] Centers for Disease Control and Prevention. Analyzing Quantitative Data for Evaluation. Evaluation Briefs. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.cdc.gov/healthyyouth/evaluation/pdf/brief20.pdf

[10] IBM. IBM SPSS Statistics. https://www.ibm.com/products/spss-statistics?utm_content=SRCWW&p1=Search&p4=43700050715561164&p5=e&p9=58700005519276965&gclid=Cj0KCQjwkdO0BhDxARIsANkNcrcI-kkOrD5XTlg1-nAd1bHkVPl66qjxbdRgW-1ot1xrxTSyYpnVzoUaAvloEALw_wcB&gclsrc=aw.ds

[11] SAS. https://www.sas.com/en_us/home.html

[12] Stata. https://www.stata.com/

[13] R. The R Project for Statistical Computing. https://www.r-project.org/

[14] Centers for Disease Control and Prevention. Analyzing Quantitative Data for Evaluation. Evaluation Briefs. chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.cdc.gov/healthyyouth/evaluation/pdf/brief20.pdf

[15] Stata. https://www.stata.com/

[16] Stata. https://www.stata.com/

[17] Stata. https://www.stata.com/

[18] Stata. https://www.stata.com/

[19] Agency for Healthcare Research and Quality. Race, Ethnicity, and Language Data: Standardization for Health Care Quality Improvement. https://www.ahrq.gov/research/findings/final-reports/iomracereport/reldata5.html