Descriptive Research

Introduction

If you lived in ancient Babylon, you would have spent more time staring at the stars than you do today. As night fell, there were no lights, no television or phone for entertainment, not even a book to read by candlelight. So, what did people do? They looked up.

Even for the average Babylonian, the heavens were significant. The calendar and many social events were based on the timing of the moon. But for a small group of scholars known as the scribes of Enūma Anu Enlil, the stars were special. Each night these scholars climbed to the top of their temples, armed with sticks for measuring angles, water clocks for tracking time, and clay tablets for writing, and they did something mundane and miraculous: they watched and recorded.

Night after night, year after year, generation after generation, these early astronomers noted the positions of stars and planets, the timing of eclipses, the appearance of comets, and the changing phases of the moon. They used their eyes and instruments to describe the heavens. Yet what made their work truly extraordinary wasn't just their observations, but their painstaking record-keeping. Each night's events were pressed into soft clay tablets, then baked hard in the sun to create permanent records. Over centuries, they amassed thousands of tablets—one of humanity's first scientific databases.

At first, these records were dull—just endless lists of events. "Venus appeared in the east this morning," one might read. "Jupiter stands in the position of the Fish constellation," said another. But with time patterns emerged. The Babylonians noticed that certain events repeated in predictable cycles. They discovered, for example, that Venus appeared as the morning star for 263 days, disappeared for 8 days, then reappeared in the evening for 263 more days. Understanding these patterns allowed the scholars to make predictions.

When would the next lunar eclipse occur? When would Jupiter return to a particular constellation? The records held the answers. And what started as an attempt to describe the stars eventually led to sophisticated theories about the entire cosmos by astronomers like Ptolemy, Copernicus, and Kepler.

What can the ancient Babylonians teach people today about describing human behavior? One lesson is that before we can predict events or explain why they happen, we must first carefully observe and describe them. This principle—that systematic description is the foundation of scientific understanding—remains just as true today as it was thousands of years ago.

In this chapter, we will learn how modern behavioral scientists conduct descriptive research. In Module 3.1, we will revisit the concept of measurement, introduced in Chapter 1. We will explain how behavioral scientists turn abstract ideas, called theoretical constructs, into numbers that can be analyzed. After discussing measurement, Module 3.2 will explore the methods behavioral scientists use to conduct descriptive research. In Module 3.3, we will use the survey tools introduced in Chapter 2 and show you how to create a descriptive study using survey platforms. In Module 3.4, we will analyze data from a descriptive study with basic statistics. Finally, in Module 3.5, you will have a chance to develop your own descriptive study, gather data, and analyze it to describe what people think about a moral dilemma. Throughout the chapter, remember the lesson of the Babylonians: careful description often reveals patterns that lead to predictions and, eventually, to deeper understanding.

Chapter Outline

Science depends upon measuring things. In chemistry, measures may be the volume, density, mass, or pH of a substance. In biology, they may be the age, growth rate, diet, or metabolic rate of an organism. In astronomy, scientists often measure light, temperature, distance, and the size of celestial bodies. And, in physics, common measures are time, speed, distance, and force. To understand the world, these measures are important.

Measures are important in the behavioral sciences, too. In psychology, scientists measure things like emotions, memory, and personality, as described in previous chapters. In political science, researchers measure which candidate voters prefer and how much people trust the political system. Economists measure how businesses feel about the economy and how much inflation consumers expect in the future. Sociologists measure patterns of social interaction and changes in community attitudes. Each of these measures helps describe what people think and do.

But measurement is often just the beginning. To truly understand people, researchers must carefully describe what they observe. For instance, a single measurement can tell how anxious a person feels right now. But if researchers systematically gather and organize many measurements, they can describe broader patterns, like how anxiety changes throughout the day, varies across situations, or differs between groups of people. Before doing this, however, the researchers need to be confident in what their measures represent.

What are Variables?

Across the behavioral sciences, researchers conduct studies to investigate people's thoughts, feelings, and behavior. Each study measures what are called variables. As you learned in Chapter 1, variables are any psychological, physical, or social characteristic that can be assigned a numerical value. When you took the TIPI, it measured the Big Five traits—openness to experience, conscientiousness, extraversion, neuroticism (or emotional stability), and agreeableness (see Box 1.1). Each of these traits is a variable.

Variables, by definition, change from person to person. For example, age is a variable. If you walk down a street in your city or town and ask each person you pass about their age, you will receive answers that vary. Some people may be born in the same year and occasionally you might find people born on the same day, but overall, you will receive answers that differ because age is a characteristic that varies across people. If your questionnaires measured personality, household income, or life satisfaction, the answers you receive would vary on these measures, too. For a characteristic to be a variable it must be capable of taking different values across people, situations, or points in time.

Variables are the building blocks of research. They are what scientists use to form hypotheses, design studies, analyze data, and interpret results. Not all variables are the same, however. Behavioral scientists distinguish between independent variables, dependent variables, and control variables. We will learn about these different types of variables in later chapters.

A variable can be contrasted with a constant. Whereas variables take different values from person to person, constants stay the same. For example, if everyone in a study was born in the United States, country of birth would be a constant.

Sometimes, researchers intentionally transform a variable into a constant. Imagine, for instance, an experiment conducted to see if feeling afraid makes people seek social connection with others. The researchers might be concerned that men and women will respond differently because of social expectations, as men are often expected to hide their fear whereas women may express it more openly. To make the study simpler, the researchers might include only women or only men (e.g., Schacter, 1959). Doing so would make gender a constant, allowing the researchers to focus on how fear affects social connection among a single gender such as women. In Chapters 5 and 6, we will examine how variables and constants can be combined within a study to understand the causes of behavior.

Conceptual vs. Operational Definitions

Across the behavioral sciences, some variables are tangible and others are abstract. Heart rate, for instance, is tangible—it can be felt, heard, and measured with various devices. Household income is also tangible; there are dollars that represent the money flowing into a household during the year. Life satisfaction, on the other hand, is abstract. There is nothing specific that we can point to and say, 'That is life satisfaction,' although sipping a Mai Tai on the beach in Hawaii might be close!

Abstract variables in the behavioral sciences are called theoretical constructs because they are difficult to observe directly. The challenge for research lies in taking a theoretical construct like anxiety, extraversion, happiness, or regret and finding a way to measure it. This process is both challenging and important because it's only through measurement that a construct can become a variable.

To conduct a scientific study, a researcher must define the variable they are measuring. An operational definition specifies exactly how the researcher will turn the concept into a quantitative measure, where each person is assigned a number that corresponds to the concept being measured.

Consider anxiety. Everyone knows what it feels like to worry. But how can a psychologist translate the subjective experience of anxiety into a precise measure where each person who is studied receives a score?

A common method is to ask questions. These questions often form a scale or measurement instrument. To measure anxiety, for example, many behavioral scientists use the Generalized Anxiety Disorder 7-item scale, abbreviated GAD-7 (Spitzer et al., 2006). The items from this instrument are in Table 3.1.

By asking people to complete the GAD-7, each person's feelings of anxiety can be translated into a number. By summing the numbers from all items within the scale, researchers obtain an overall score that can range from zero (no anxiety) to 21 (severe anxiety). This total score is the operational definition of anxiety.

In Chapter 1, we encountered the TIPI, which is an operational definition of the Big Five personality traits. The TIPI consists of specific questions that participants answer about themselves. By combining responses to these questions, researchers generate numerical scores for each of the Big Five traits. These scores serve as the operational definitions of the personality traits, just as a GAD-7 total score serves as the operational definition of anxiety.

Psychometrically, there is nothing special about the GAD-7 or anxiety. Behavioral scientists have developed several ways to measure anxiety (e.g., Newton & Buck, 2000; Nomura et al., 2006; Shelton et al., 2010; Vigil-Colet et al., 2008) and thousands of ways to measure other psychological constructs. Each measure follows the same process: take a behavior or experience, create an instrument to assess it, and assign each person a number. These numbers are operational definitions that form the basis of behavioral research. In the next chapter, you will learn how to create measurement instruments for virtually any construct you would like to measure.

What is Descriptive Research?

Descriptive research is a type of study that systematically measures and records how frequently something happens in a group of people. When we talk about frequency, we mean how often a behavior or attitude occurs or how data points are spread out on a graph, like in a bell curve.

Recall, as an example, the measure of anxiety introduced in Module 3.1, the GAD-7. In a descriptive study, researchers could give people the GAD-7 and use their scores to describe what percentage of people have low, moderate, or severe anxiety. They could also plot the scores to see how anxiety is distributed across people. In other situations, behavioral scientists might want to describe how common different types of cancer are, how much credit card debt the typical household carries, or even how fast people run a marathon. Each of these questions can be answered with descriptive methods.

Describing people's thoughts and behaviors can be especially valuable because it often leads to more complex questions about when and why something occurs. The research project in Module 3.3 offers a good example of this. But for now, let's look at a few examples of descriptive research and highlight how they help researchers understand human behavior.

Marathon Finishing Times

A marathon is a race that lasts 26.2 miles. How long do you think it takes people to run that far?

Fortunately, this question doesn't require a guess. Marathon times are something that can be measured with a variation of the performance tools we introduced in Chapter 2 (a computer chip often tracks when people cross the start and finish lines). Figure 3.1 (left panel) shows the marathon times for almost 10 million runners (Allen et al., 2016). How does this data characterize people's behavior?

First, the data allows us to look at the extremes, reporting both how fast and how slow people run. For example, the fastest runners finished in under 2 hours and 30 minutes—less than six minutes per mile!—while the slowest runners took 7 hours, or about 16 minutes per mile.

Second, the data allows us to describe what is most typical or common. Using descriptive statistics, we could calculate the sample average to describe how long it takes most people to run a marathon. Similarly, looking at the figure shows that the highest line is at the four-hour mark, with more than 100,000 people achieving this time. Most other times are between 3 hours and 5 hours, giving us a sense of where most people finish.

Third, plotting the times allows us to see the overall distribution. Marathon times, it turns out, are close to a normal distribution. A normally distributed variable is one in which most values cluster around the middle and fewer values appear toward the tails, or sides, of the figure. This distribution is often referred to as a 'bell curve' (see Figure 3.1, right panel). Regardless of how a variable is distributed, characterizing the frequency of a behavior is an important function of descriptive research.

Mental Health Among U.S. Adults

A very common method for describing behavior is survey research. Within the United States, for instance, the National Institutes of Mental Health has a mandate to promote awareness of mental illness and to guide policy for treatment. To fulfill this mandate, the agency needs to know how many people suffer from mental illness—a question for descriptive research.

During 2021, the National Institutes of Mental Health conducted a large survey to understand what percentage of U.S. adults experienced some form of mental illness throughout the year. To conduct this study, the researchers compiled a list of every residential address in the country (NIMH, 2023). Then, they selected households to participate. The data collection stretched across an entire year and ended with more than 70,000 interviews!

Figure 3.2 shows the results. As seen in the left-most bar, the overall rate of mental illness was 22.8%. Across social groups such as sex, age, and race, the rate of mental illness varied. Women, for instance, were more likely to experience mental illness than men, younger people were more likely to experience mental illness than older people, and White people were more likely to experience mental illness than African American, Asian, or Hispanic people.

While the survey allows researchers to characterize the frequency of mental illness among U.S. adults, it also does at least two other things. First, it reveals trends and patterns. In the data, we can see that age is associated with mental illness; younger people report mental illness at twice the rate of older people. Similarly, women report more mental illness than men. These patterns suggest associations that can be examined in future studies.

Second, the data establishes a baseline, or norm. In future studies, researchers can track whether mental health improves or declines among different groups over time. Ideally, when the data show an unexpected jump in mental illness among specific groups, policy makers and mental health care providers could respond with appropriate resources (National Survey on Drug Use and Health, n.d.). In this way, descriptive research can make society more responsive to emerging problems.

Water Usage and Behavioral Sensors

Descriptive data can come from performance measures, surveys, or observations. Thanks to technology, many observations can be made indirectly, like in a study where researchers wanted to know how much water people use in the shower (Pereira-Doel et al., 2024).

Water use is a growing concern in many countries because people are depleting the supply of groundwater (Rojanasakul et al., 2023). Heating water for showers also consumes energy that is thought to contribute to climate change. Yet, before proposing solutions to either of these problems researchers wanted to describe people's showering habits.

To do that, they installed sensors—a device that passively records input from the physical environment—in hundreds of showers around a university campus. The sensors recorded how long people showered over 39 weeks, resulting in 86,000 observations (the sensors only recorded water usage to protect people's privacy). From this data, the researchers calculated that the average shower lasted 6.7 minutes (Figure 3.3). However, the sensors also measured the water pressure of each shower and how much water was used. From the descriptive statistics, the researchers were able to tell that higher water pressure was associated with shorter showers. From this pattern, the researchers developed a hypothesis for future research—higher water pressure should cause people to take shorter showers—and were able to make recommendations to guide water use policy in the future. Thus, as we will see throughout this book, descriptive research is often the starting point of further and more in-depth exploration.

Now that we have seen a few examples of descriptive research, the next module will provide the opportunity to walk through a guided project, programming a survey to collect descriptive data, analyzing the results, and thinking about what they mean.

Let's review what we have learned so far. First, measurement is important. For any study, researchers must define not only what they are interested in but how they will measure it. Second, descriptive research gathers data to characterize what people think or do. Good descriptive research allows us to identify patterns and forge hypotheses for new studies. While these concepts illustrate the elements of descriptive research, they do not show how a study comes together or how researchers work with descriptive data. Therefore, we will walk through an example project from start to finish.

The example project examines how people respond to a moral dilemma. Working on this project will draw on many of the things we have covered in previous chapters. For example, after reading about the moral dilemma below, you will have the chance to think about your response and to create a hypothesis about how others will respond. Next, we will walk through how people's judgments are measured in the dilemma and then program the survey together in Qualtrics. As described in Chapter 2, survey platforms like Qualtrics are the primary tool behavioral scientists use to gather self-reported data.

After programming the survey, we will explain how the authors of this book used the same dilemma to gather data from 200 online participants. Together, we will download the data file and conduct basic descriptive analyses that describe people's reactions to the dilemma. After considering the implications of the study, you will be ready to modify the scenario and conduct your own descriptive project in Module 3.5.

Describing People's Reactions to a Moral Dilemma

In the 1950s, psychologist Lawrence Kohlberg began studying how people reason about right and wrong. To conduct his studies, Kohlberg presented people with dilemmas that pit different moral principles against each other. His most famous scenario, known as the Heinz dilemma, asks people to weigh the sanctity of human life against the law.

Here is a modern version of the dilemma:

A woman is on her deathbed, and the only drug that doctors say could save her is a newly discovered form of medication. This drug is expensive to produce. The company that manufactures the drug also sells it at a significant markup. Each dose of the medication costs the company $20,000 to produce, but they sell it for more than $200,000 per dose.

The woman's husband, Heinz, exhausts all his options trying to gather the money. He asks family and friends, he starts a GoFundMe, he takes a second job, and he appeals to the company for a discount. Altogether, he only manages to collect about $100,000—half of what the drug costs.

Heinz pleads with his insurance and the company that makes the drug, explaining his wife's dire condition and asking for a discount or a payment plan, but everyone refuses. Faced with the imminent death of his wife and unable to afford the drug legitimately, Heinz feels compelled to break into a pharmacy and steal the medication.

Choosing What to Measure

When conducting descriptive research, behavioral scientists must decide exactly how to measure the concepts they are interested in. For the Heinz dilemma, we want to understand two aspects of people's reactions: what they think Heinz should do and how morally acceptable they find his actions.

For the first aspect, we can use a simple yes/no question: "Should Heinz have broken into the pharmacy to steal the drug for his wife?" This question forces people to decide, just as Heinz had to do.

However, people's views might be more nuanced. Someone might think Heinz shouldn't steal the drug but still see stealing the drug as somewhat justified, given the circumstances. To examine this, we can use a second question: "How morally acceptable was it for Heinz to steal the drug?" Participants can answer using a 7-point scale ranging from "Not at all acceptable" to "Entirely acceptable."

Once the measures for a study have been selected, the next step is to program them on a survey platform. In the activity that follows, we will show you how to program the Heinz dilemma.

Step 1: Setting Up Your Project

First, login to your Qualtrics account by using the credentials your instructor provided or creating a free student account: https://www.qualtrics.com/free-account/.

After logging in, click the "Create Project" button. Select "Survey" from the options provided, then choose "Blank Project." Name your survey "Heinz Dilemma Descriptive Project" and click "Create Project."

You will now see the Qualtrics Survey Builder (Figure 3.4). Notice the menu on the left side showing icons for the "Builder" "Survey Flow," "Look and Feel" and other options. Also notice the tabs across the top for things like "Distributions," "Data & Analysis," and "Reports." For now, we will build the questionnaire on the survey tab.

Step 2: Organizing the Survey into Blocks

Professional researchers organize their surveys into logical sections called "blocks." For this study, we will create three blocks:

1. Introduction and Consent
2. Heinz Dilemma
3. Demographic Questions

Look for the "Default Question Block" that appears in your survey. Click on the block title to rename it "Introduction and Consent." Then click the "Add Block" button at the bottom to create two more blocks. Name them "Heinz Dilemma" and "Demographic Questions." The result should look like Figure 3.5.

Step 3: Creating the Introduction

In the Introduction block, click "Add Question." Choose "Text/Graphic" from the question types. In the box that appears for editing, enter the text below to introduce your study:

"Welcome to the Perceptions of Social Events Study. This study examines how people respond to a social situation. Your participation will involve reading a scenario and answering a few simple questions. It will take approximately 3 minutes to complete. Your responses will be anonymous and used for research purposes only. Please make sure to complete the study in one sitting and in a quiet location without distraction.

By clicking 'Next' below, you indicate that you are at least 18 years old and voluntarily agree to participate in this study. You may exit the survey at any time if you wish to discontinue participation."

Add another question and choose multiple choice. Using the menu on the left, limit the answer options to two and paste the content below into the question text:

"By clicking 'Continue' below, you indicate that you are at least 18 years old and voluntarily agree to participate in this study. You may exit the survey at any time if you wish to discontinue participation."

Finally, edit the two answer options to read: 1) "Continue with the study", and 2) "Exit the study."

Within the survey, this information will look like Figure 3.6. If you were actually gathering the data for this project, the consent form would be longer (see Chapter 15) and you would need to ensure people who select "End participation" are taken to the end of the survey. We will ignore these details for now.

Step 4: Adding the Dilemma and Outcome Measures

In the Heinz Dilemma block, add the dilemma participants will read. Click "Add Question" and select "Text/Graphic" from the question types. Then paste the Heinz dilemma from earlier in the chapter into the question box. You may need to create spaces between paragraphs to ensure the dilemma is easy for participants to read.

The result should look like Figure 3.7 in the survey builder. When the survey is presented to participants it will appear as in Figure 3.8.

After the dilemma, enter a "page break." Page breaks divide the content within a block so that participants see each question on a separate screen. After the page break, add a multiple-choice question. For the question text write: "Should Heinz have broken into the pharmacy to steal the drug for his wife?" Then, edit the number of answer options down to two, and write "yes" and "no" into the answer options. The result should look like Figure 3.9.

After the yes/no item, add another multiple-choice question. Paste the question, "How morally acceptable was it for Heinz to steal the drug?" into the question text. Then, edit the answer options so that there are 7 choices ranging from 1 "Not at all acceptable" to 7 "Entirely acceptable." The result should look like Figure 3.10.

Now, you are ready to move on to the last block of the survey.

Step 5: Adding Demographic Questions

In the Demographic questions block, you can add questions that gather information about who is participating in the study. These include things like people's age, gender, race, and education level.

The easiest way to add these questions is with Qualtrics's question library. This library will not be appropriate for every study, but for this study you can click the "Import from library" button within the demographics block. In the side window that appears, select "Demographics" under the Qualtrics certified header, and then choose "Global demographics." The first question should be about age. If you click on the item, you will see a preview of how it appears and a button to "Insert question" (Figure 3.11). Add the question.

If you explore the Global and U.S. Demographics tabs, you should find questions asking about participant's gender, race, education level, and household income. Add each of these to the demographics block, and then the survey is done.

Step 6: Previewing Your Survey and Generating the Survey Link

Once you have built the survey, you should preview it. Previewing allows you to see how the survey appears to participants and to make sure that everything looks professional and works as expected.

To preview, click on the "Distributions" tab in the menu. On the following page, select the option for "Anonymous link." You should see something that looks like Figure 3.12. You can copy this link into a new browser to preview the survey. Later, you can also use this same link to send the survey to participants or paste it into a participant recruitment site like Connect.

Congratulations! You have just created a survey that can be used to describe people's reactions to the Heinz dilemma. In creating this survey, you have seen how researchers use survey platforms like Qualtrics and Engage to organize studies into logical blocks, add stimuli like the moral dilemma, and program different types of questions (text, matrix tables, multiple choice). You have also seen how a survey platform that looks a bit technical on the back end can render clean and professional-looking surveys that guide participants through a research experience.

Gathering Descriptive Data Matched to the U.S. Census

Normally, the next step after programming a survey is to gather data. We have done that for you.

We gathered data from 100 participants on Connect. Before launching the survey, we had to consider several practical details like how long the study would take and how much to pay participants. Through pilot testing, we found that most people completed the survey in 2-3 minutes. Following standard payment practices, we offered 50 cents for taking the study—equivalent to about $12 per hour.

In addition to payment, we had to consider who to sample—an important decision in any descriptive research project. We wanted to describe how people in the U.S. react to the Heinz dilemma, so we needed participants who represented the demographics of the U.S. population. Connect offers a feature called Census Match, which creates quotas that helped achieve this goal. The Census Match quotas specified what percentage of participants should come from different demographic groups.

We created quotas to match the U.S. Census numbers for age, gender, race, and ethnicity. For example, about 13% of the U.S. population identifies as Black or African American. Using the quotas, we ensured 13% of our sample did too. Quotas matched to the U.S. Census resulted in a sample that was reasonably matched to key demographics of the U.S. population. Chapter 9 discusses issues of sampling and representativeness in more detail. Within ten minutes of launching the study, we had responses from 100 participants. In the next module, we will show you how to analyze this data, before giving you the chance to collect your own descriptive data at the end of the chapter.

Downloading the Heinz Dilemma Data From OSF

As mentioned in Chapter 2, the materials for each project in this book are stored on the Research in the Cloud OSF page. You can access that page here: https://osf.io/a8kev/.

When you arrive on the OSF page, click the "Files" option in the menu (Figure 3.13). Then, choose the folder labeled "Ch. 3 – Descriptive Research." Inside, download the file named "RITC_DATA_CH03_HeinzDilemma.sav" (Figure 3.14). All files in this book follow this naming convention. RITC stands for Research in the Cloud—the name of this book. Then, the first word describes the file type—DATA, SURVEY, SYNTAX, or MATERIALS—followed by the chapter number and a description of the project.

If you are using SPSS, download the .sav data file; if you're using another statistical program, choose the .csv file. Once the file is open in your analysis program, you are ready to analyze the data.

For the first measure—whether Heinz should steal the drug—we simply calculated the percentage of participants who answered "yes" versus "no." The results revealed an interesting split: only 47% of participants thought Heinz should steal the drug. This suggests people were divided about Heinz's choice, with a slight majority feeling he should not break the law even to save his wife's life.

However, the results from the second measure showed a different pattern. When people were asked how morally acceptable it was for Heinz to steal the drug on a scale from 1 (not at all acceptable) to 7 (entirely acceptable), they gave an average rating of 4.39 (Figure 3.15). Given that 4 indicates neutrality, the data showed that people leaned toward feeling that Heinz's actions were morally acceptable. Indeed, the most common response (the mode) was 7, and 54% of people selected a response showing some kind of approval (5 or above). By contrast, 36% of people selected a score showing disapproval (3 or below). The other 10% were in the middle. Overall, more people saw Heinz's action as morally justified even though the majority of people said he should not have stolen the drug.

Putting Things Together: The Heinz Dilemma and the Scientific Process

The data from the Heinz dilemma show that when people are asked whether it's okay for Heinz to steal the drug most said 'no.' But when people judged the morality of the action, most indicated that Heinz's action was morally acceptable. If this data is representative of U.S. adults, we can characterize people's reactions to the dilemma as ambivalent. Most people are against stealing but find it somewhat acceptable in this case. Characterizing people's thoughts and behaviors, you may remember, is one of the outcomes of descriptive research.

Another outcome of descriptive research is suggesting hypotheses and directions for future research. In this case, an avenue for future research is clear: investigating why there is a difference in people's answers depending upon whether they are judging whether Heinz should have stolen the drug (most people said no) and whether stealing was morally acceptable (most people say yes). The data from this study cannot explain why people's responses were mixed, but it can help us form a hypothesis and come up with ways to test that hypothesis in future research. For instance, the pattern may suggest that many people were influenced not just by sympathy for Heinz's situation but also by a concern for rules and societal order. While participants recognized the importance of saving a life, they ultimately prioritized respect for law and order. In a future study, we might hypothesize that a concern for social order or the ability to take Heinz's perspective would influence people's judgments.

Finally, you may recall that the process of forming hypotheses, gathering data, adjusting ideas, and forming new hypotheses for further study is exactly how the scientific process is supposed to work (see Figure 1.6 from Chapter 1). Even a simple moral dilemma with just two questions can reveal interesting patterns or contradictions that are worthy of further investigation. In later chapters, we will undertake these investigations together, but for now it is important to remember that science is a cycle that leads to an ever-increasing understanding of people and the world.

Gathering Descriptive Data Online

The video we created for this assignment will guide you through creating a study on Connect: https://bit.ly/Ch3_SHD. To demonstrate some of the steps below, we will use the Heinz Dilemma survey you created earlier in the chapter.

To get started, login to the researcher side of Connect, just as you did in Chapter 2. Navigate to: https://account.cloudresearch.com/Account/Login, and choose "Connect for Researchers." Login with the same email and password you used previously.

When you arrive at the dashboard, click the blue "Create Project" button (Figure 3.16). You will be ready to gather data in five simple steps.

Step 1: Basic Information

The first thing to do is provide some basic information such as a title and a description of what participants will be asked to do (Figure 3.17). Give your project the title "Social Events Survey" and describe it as "A survey about a social situation."

Within the basic information section, you will also see checkboxes to limit the devices participants can use to take the study or to communicate whether the study requires audio, a camera, or downloading software to participate. You can leave these settings alone for this project. Then, you are on to step two.

Step 2: Project Link

In Step 2, you provide the link to your project (Figure 3.18). Any third-party URL will work. In this case, paste the "anonymous link" from your Qualtrics survey (see Figure 3.12 if you need a reminder of where to find this link).

Step 3: Cost

In Step 3, you decide the details that determine how much your project will cost: how many participants you want to sample and how much you will pay them. You also need to estimate how long it will take each person to complete the project.

For this example, the recommended settings are 100 participants, with each paid 50 cents for a 3-minute study (Figure 3.19).

Step 4: Participant Targeting

In Step 4, you specify any demographic criteria participants must meet. Connect has hundreds of data points about each participant. You can recruit people based on these characteristics.

Because this study seeks to describe people's reactions to a moral dilemma, you will want to add a Census Matched template. To do that, select the demographic targeting option. When the popup appears, choose that you want participants from the United States and then click "Apply Census Matched Template" (Figure 3.20). The system will ask you to confirm your choice, and when you do, the quotas will automatically appear in proportion to the U.S. population. Click next and apply to confirm your selections. You are done with Step 4.

Step 5: Completion

The last step in creating a project is determining how participants will end it. The easiest option is to use a redirect URL. A redirect URL sends anyone who completes your Qualtrics survey back to Connect. These participants get marked as "complete" and stay in a pending status for 14 days until you approve or reject their submission.

To configure a redirect URL, you need to copy the URL link provided in Connect (Figure 3.21). Then, you need to paste this link into Qualtrics. Navigate to the bottom of your Qualtrics survey and click on the "End of survey" block. A side panel will open to the left where you can change the "End of survey message" from "Default" to "Redirect to URL" (Figure 3.22). Finally, paste the URL from Connect into the website URL box. Your study is all set.

Select "Save changes." The project will appear as a draft on your dashboard where you can launch it if your idea is chosen in class.

Summary

Throughout this chapter, you have explored the essential components of descriptive research. You began by examining measurement, learning how behavioral scientists transform abstract concepts like anxiety or moral reasoning into quantifiable variables through operational definitions. Then, you explored various methods of descriptive research, from large-scale surveys like the National Survey on Drug Use and Health to observational studies using environmental sensors. Each approach provides a different window into human behavior.

Next, you gained practical experience by working with the Heinz dilemma, learning to program a survey in Qualtrics, analyze data in SPSS, and interpret findings about moral reasoning. The discrepancy between people's binary choices and their ratings of moral acceptability reveals the complexity of human judgment—a complexity that can only be uncovered through careful descriptive research.

Finally, you created your own descriptive study by modifying the Heinz dilemma, generating hypotheses, and preparing to collect data from a real source of research participants. This process mirrors what professional researchers do every day.

Remember that while description is important it is rarely the end goal. Like the Babylonian astronomical records that eventually led to sophisticated theories about the cosmos, today's descriptive research lays the groundwork for tomorrow's explanatory theories. The patterns scientists observe in data often raise new questions: Why do these patterns exist? What causes them? How might they be changed?

As you continue your journey through this book, you will build upon the foundation of descriptive research to explore correlational and experimental methods that help answer these deeper questions. But don't forget that it all begins with careful description—the simple yet profound act of systematically observing and recording what people think, feel, and do.