What best type of data are you collecting?
The best type of data are you collecting
The type of data are very important. Mixed methods research uses quantitative and qualitative information in order to answer a research question. Quantitative research involves the analysis of numerical data that can be used for statistical analysis, while qualitative research involves collecting data for exploratory purposes or to find common themes.
“Data is the new oil.” Today data is everywhere in every field. Whether you are a data scientist, marketer, businessman, data analyst, researcher, or you are in any other profession, you need to play or experiment with raw or structured data. This data is so important for us that it becomes important to handle and store it properly, without any error. While working on these data, it is important to know the types of data to process them and get the right results. There are two types of data: Qualitative and Quantitative data, which are further classified into:
The data is classified into four categories:
- Nominal data.
- Ordinal data.
- Discrete data.
- Continuous data.
So there are 4 Types of Data:
Mixed Methods Research?
Definition is a design that utilizes both quantitative (numeric) and qualitative (descriptive) research elements to gain a conclusion for a study. A mixed methods study is likely to utilize several data collection practices and evaluation processes that are appropriate for both quantitative and qualitative approaches. Mixed methods research can most easily be identified by the following characteristics:
- Collection and analysis of both quantitative and qualitative data
- Integration of the data during the collection process
- Theoretical model(s) serving as a research framework
A mixed methods research design is most often used when there is an interest both in testing theories or integrating a specific theoretical perspective while also exploring a better understanding of an experience or process. If the research question cannot be answered just by analyzing qualitative or quantitative data alone, it would be best to use a mixed methods approach. This type of research is most often done in behavioral and health settings since both usually involve completed situational research from a large sample size.
Mixed Methods Research Examples
The following constitutes two different types of mixed methods research examples:
A healthcare researcher wants to investigate rising healthcare costs for patients at two different hospitals who are receiving different treatments for a specific illness. The researcher decides that they want to look at the trend of costs from 10 years to the current year, insurance premiums, and the out-of-pocket costs to patients. They interview doctors from both hospitals as a means to inform of any changes in treatment over the past 10 years and to justify the rising costs. The first phase constitutes the quantitative assessment, and the second phase constitutes the qualitative portion of the explanatory mixed methods study.
A political researcher is interested in studying the trends of voters under the age of 35 in a specific region. He wants to study why the voter turnout is so low, while also showing the trend of overall voters over the past years in the region. He decides that an embedded mixed methods study would be best and starts by interviewing voters under the age of 35 and attending voter registration drives to collect information. He also gathers voting trends from the local clerk of court and compares them among pre-selected age ranges. He is able to give supporting quantitative data when hypothesizing why voter turnout was so low among the 35 and under registered voters.
Mixed methods research uses quantitative and qualitative information in order to answer a research question. Quantitative research involves the analysis of numerical data that can be used for statistical analysis, while qualitative research involves collecting data for exploratory purposes or to find common themes. Mixed methods data should be used when the research question cannot be answered by exclusively using either qualitative or quantitative data collecting methods. Strengths of mixed methods designs include that the data is richer, and the data collection methods are more flexible. Weaknesses of mixed methods research are that it can be very time-consuming and complex.
There are four major designs of mixed methods research: exploratory, explanatory, convergent, and embedded. When conducting any type of research regarding the rate of recurrence of activity or phenomenon, it is important that the researcher includes operational definitions. For example, if a researcher wants to determine the amount of aerobic activity among students, the operational definition of what is considered aerobic activity should be given.
Operational definitions can be limiting, however, because they restrict the reader’s view about what can or cannot be perceived as an appropriate variable in the study. Mixed methods research allows readers to make generalizations that can be used among similar events or circumstances. Mixed methods research is most commonly used in healthcare and medical science; however, it can also be used in social science settings.
Are you going to do market research and don’t know what data collection technique you are going to use? I remind you that the design of your research will depend on this, so think carefully before saying whether you will do interviews, use the observation method or perhaps online surveys.
Before deciding which method you will choose to collect data, it is important to know what you want to obtain through this research, to be clear about the objectives to know which data collection technique will give us the best results.
What is data collection?
Data collection refers to the systematic approach of gathering and measuring information from various sources in order to obtain a complete and accurate picture of an area of interest.
Collecting data allows an individual or company to answer relevant questions, evaluate results, and better anticipate future probabilities and trends. Accuracy in data collection is essential to ensure the integrity of a study, sound business decisions, and quality assurance.
For example, you can collect data through mobile apps, website visits, loyalty programs, and online surveys to learn more about customers.
How to collect data correctly?
There are different data collection methods that can be useful to you. The choice of method depends on the strategy, type of variable, desired precision, collection point, and interviewer skills.
Types of data: input metrics and output metrics
Generally speaking, you’ll want to collect both input metrics and output metrics to achieve your objective.
Metrics are measurements that are used to assess and compare. Words typed per minute is a metric, because you can measure it and use it to assess and compare people’s typing skills. Input metrics you can directly control. How often you water your plant is an input metric. Output metrics you cannot directly control. How much your plant grows is an output metric.
Inputs create outputs, and outputs are what you care about. You want your plant to grow, so you water it. You want to include input and output metrics in your data collection because the output will tell you if you got far enough, and if you haven’t, you can turn the knobs on your inputs to get there.
Types of data: quantitative and qualitative
- Both quantitative and qualitative data are useful to make decisions.
- Quantitative data is expressed in numbers. It tells you what is happening.
- Qualitative data is expressed in words. It often tells you why it’s happening.
- Some time ago I ran a product line that allowed small businesses to accept credit card payments from customers.
- A funny thing happened one month. Once we enrolled a new business, their credit card payments would start off strong then suddenly stop after a few days.
- We had the quantitative data that there was something wrong, and it rang the alarm bells to take action.
- But the data we had wasn’t pointing us in any direction, so we didn’t know what action to take.
- We set about calling customers and collecting qualitative data. We asked them in a friendly yet direct way why they had suddenly stopped using our product.
- Collecting qualitative data from customers empowered us to take action on the quantitative data that something was wrong.
- They told us that they were sold on getting a free iPod if they ran $2,000 in payments using our product.
- You see, we incentivized small businesses to work with us by offering them the hardware they needed for free.
- I realized what went wrong. We recently had changed the commission structure for salespeople.
- We offered a bonus to salespeople if their merchants cleared $2,000 in credit card payments in the first week.
- So instead of selling merchants on our value proposition, our small sales team sold a free iPod if the merchant cleared the payments number.
- In this case, qualitative data helped us uncover dubious behavior and enabled us to revamp our sales team. It turns out qualitative data is often this useful.
Examples of quantitative data
- Revenue and costs ($10K in sales this month but $12K in costs)
- Ratings (7 out of 10, or 4 out of 5 stars)
- Length of time (Delivery method #1 takes 2 days to ship and method #2 takes 5 days)
- Price (Product A costs $7.99 and Product B costs $10.99)
- Quantity (Cooking method #1 produces 4 servings and method #2 produces 10 servings)
Examples of qualitative data
- Product changes (Product enhancement launched last month which increased costs but not revenue)
- Reviews (“Restaurant had good food and service, but the wait was long so I gave it 4 out of 5 stars”)
- Interviews or comments (“I’d rather save money on shipping and wait a while, so I chose the longer delivery method”)
- Features (Product B comes with a lifetime warranty but Product A doesn’t)
- Observations (Cooking method #2 has less consistency in quality than method #1)
Data collection methods and tools
By now you’ve done the following:
- Defined a question to answer with the data
- Decided what types of data are needed based on the question
You’re ready to explore the best ways to collect the data. Let’s begin.