Understanding Quantitative Research Approaches

Defining quantitative research

The systematic approaches that ground quantitative research involve hundreds or thousands of data points for one research project. The wonder of quantitative research is that each data point, or row in a spreadsheet, is a person and has a human story to tell. 

Quantitative research aggregates voices and distills them into numbers that uncover trends, illuminate relationships, and correlations that inform decision-making with solid evidence and clarity.

 

So why choose a quantitative approach? 

Because you want a very clear story grounded in statistical rigor as a guide to making smart, data-backed decisions. 

 

Quantitative approaches shine because they:

Involve a lot of people

Large sample sizes (think hundreds or thousands) enables researchers to generalize findings because the sample is representative of the total population.  

Are grounded in statistical rigor

Allowing for precise measurement and analysis of data, providing statistically significant results that bolster confidence in research.

Reduce bias

Structured data collection and analysis methods enhance the reliability of findings. 

Boost efficiency

Quantitative methods often follow a qualitative phase, allowing researchers to validate findings by reporting the perspective of hundreds of people in a fraction of the time. 

Widen the analysis’ scope

The copious data collected in just a 20-minute (max) survey positions researchers to evaluate a broad spectrum of variables within the data. This thorough comprehension is instrumental when dealing with complex questions that require in-depth analysis. 

 

Quantitative approaches have hurdles, which include:

Limited flexibility

Once a survey is fielded, or data is gathered, there’s no opportunity to ask a follow-up question live. While it is possible to follow-up with the same people for two surveys, the likelihood of sufficient responses is small. 

Battling bots

One of the biggest concerns in data quality is making sure data represents people and not bots (an important and tantalizing topic best left for another post). The Sound combats bots by integrating various data checks within surveys and validating the checks before any analysis begins.

Missing body language cues

Numbers, words, and even images lack the cues that body language gives off. Unlike in a qualitative focus group, where one might deduce that a person is uncertain of an answer, in quantitative research, a static response is what the researcher works with.

Missing body language cues

Numbers, words, and even images lack the cues that a researcher could pick up on during an interview.

 

Getting to know the quantitative way

Quantitative approaches approach research from the same starting point as qualitative approaches—grounded in business objectives with a specific group of people to study. 

Once research has kicked off, the business objective thoroughly explored, and the approach selected, research follows a general outline: 

Consider what data is needed

Think about what type of information needs to be gathered, with an approach in mind. While most quantitative research involves numbers, words and images also count.

• Numbers: Yes, the stereotypical rows of numbers in spreadsheets. Rows that capture people’s opinions and attitudes and are coded to numbers for comparative analytics. Numerical analysis is used for everything from descriptive statistics, to  regression/predictive analysis. 
• Words: Text analysis employs a machine learning model to identify sentiment, emotion, and meaning of text. Often used for sentiment analysis or content classification, it can be applied to single-word responses, elaborate open-ends, reviews, or even social media posts.
• Images: Image analysis extracts meaningful information from images. A computer vision model that takes images as inputs and outputs numerical information (e.g., having a sample upload their favorite bag of chips, and yielding the top 3 brands of chips).

Design a survey

To capture all forms of data required for the approach to address the objective. During this process, different pathways could be written to get a dynamic data set (capturing opinions that derive from various lived experiences). Survey logic is also written to provide a smooth UX experience for respondents.    

Ready the data

The quality of quantitative research rests heavily on the quality of data. After data is collected (typically by fielding a survey or collecting already-existing data…more on that in a bit), it’s time to clean the data. 

Dive into analysis

Now that you have a robust database (including numbers, words, or images), it’s time to listen to the story that the data tells. Depending on the research approach used (more on that later), advanced analytics come into play to tease out insights and nuances for the business objective. 

Tell the story

Strip the quantitative jargon and convey the insights from the research. Just because it’s quantitative research does not mean the results have to be told in a monotone drone with a monochrome chart. Answer business objectives dynamically, knowing that research is grounded in statistically sound information.

 

Primary or secondary: research methods options

The two methods that encompass quantitative approaches are primary (collecting data oneself) and secondary (relying on already existing data).

Primary is primarily used 

Most research involves primary data collection—where the researcher collects data directly. The main approach in primary research is survey data collection.  

Question types in surveys

Span various measurement scales (nominal, ordinal, interval, and ratio) using a mix of question types (single and multi-choice, scales, matrix, or open-ends). 

Analysis methods

Custom surveys (which The Sound creates for each client) yield great data for a variety of methods in market analysis. Here are a couple favorites: 

• Cross-tabulation: Used to uncover insights that might not be obvious at first glance. This analysis organizes data into categories, revealing trends or patterns between variables. 
• Sentiment analysis: Used to sift through text to gauge emotions, opinions, and attitudes. This method helps understand perception, fine-tune strategies, and effectively respond to feedback.
• Market sizing: Used to map out the dimensions of a market. By calculating the total potential demand for a product or service in a specific market, this method reveals the scope of opportunities needed to make informed decisions about investment and growth strategies. 
• Conjoint Analysis: Used to uncover what people value most in products or services. It breaks down features into bits and pieces and asks people to choose their ideal combo. By analyzing these preferences, this analysis reveals the hidden recipe for customer satisfaction.
• Job-To-Be-Done: Used to understand the underlying human motivations that drive people to act. People are multifaceted and experience a myriad of situations each day—meaning that a brand’s competition isn’t limited to in-category. 
• Segmentation: Used to identify specific cohorts within a greater population. It groups people with similar characteristics, behaviors, or needs together. This method helps tailor products or services to specific groups, boosting satisfaction and sales.

Statistical rigor

Regardless of method, a quantitative approach then enables researchers to draw inferences and make predictions based upon the confidence in the data (looking at confidence intervals, margin of error, etc.)

 

Let’s not forget secondary research

By accessing a wide range of existing information, this research can be a cost-effective way to gain insights or can supplement primary research findings. Here are popular options:

Government sources

Can be extremely in-depth, range across multiple industries and markets, and reflect millions of people. This type of data is often instrumental for longitudinal or cultural trends analysis. 

Educational institutions

Research universities conduct in-depth studies on a variety of topics, often aggregating government data, non-profit data, and primary data. 

Client data

Includes any research that was conducted for or by companies before the present research project. Whether it’s data gathered from customer reviews or prior quantitative work, these secondary resources can help extend findings and detect trends by connecting past data to future data.

 

In closing…

Quantitative research approaches are so much more than ‘how much’ or ‘how many’—they reveal the why behind people’s actions, emotions, and behaviors. By using standardized collection methods, like surveys, quant instills confidence and rigor in findings.  

Have some research questions you’d like answered? Quantitative or qualitative, we’ll find the best approach for you. We’re here to help.