Question

The article "Super Bowls: Serving Bowl Size and Food Consumption" (Journal of the American Medical Association [2005]: 1727-1728) describes an experiment investigating how the size of serving bowls influences the amount a person eats. In this experiment, graduate students at a university were recruited to attend a Super Bowl party. The paper states that as the students arrived, they were "led in an alternating order to one of two identical buffet tables on opposite sides of an adjoining room. The tables had identical amounts of snacks, such as nuts, pretzels, and chips. All of the snacks contained approximately the same number of calories per gram. On one of the tables the snacks were set out in large serving bowls and on the second table the snacks were set out in smaller serving bowls. The students were given a plate and invited to serve themselves before going to another room to watch the game. When they arrived at the game room, their plates were weighed and the number of calories in the food on the plate was estimated." The researchers concluded that serving bowl size does make a difference, with those using large serving bowls tending to take more food. a. Do you think that the alternate assignment to the experimental groups (large serving bowls, small serving bowls) based on arrival time is "close enough" to random assignment? That is, do you think it would tend to create comparable experimental groups? b. In this study, the research assistant who weighed the plates and estimated the calorie content of the food on the plate was blinded as to which experimental group the plate belonged to and was also blinded as to the purpose of the experiment. Why do you think the researchers chose to incorporate this type of blinding?

Short answer

Alternate assignment to the experimental groups based on arrival time is not 'close enough' to random assignment because it might introduce confounding variables such as hunger level with arrival time. Blinding was incorporated in this study to reduce bias. If the assistant knew which group the subject was in and what the experiment was about, their bias might affect the measuring and interpretation of the results.

Step-by-step solution

Understanding Random Assignment

Random assignment is a fundamental part of creating a scientifically valid experiment. By randomly assigning subjects to different experimental groups, one can help to ensure that any observed differences are due to the variable being tested rather than extraneous factors. However, in this scenario, subjects were assigned to the large or small serving bowl groups based on arrival time, not random assignment. There may be underlying variables correlated with arrival time that could also influence the amount of food taken, such as hunger level.

Understanding Blinding in Experiments

Blinding is a technique used in research to eliminate bias. In this study, the research assistant who weighed the plates and estimated the calorie intake was blinded to which group the plate belonged to and to the purpose of the experiment. Blinding the research assistant in this way ensures that their knowledge does not influence the measurement and interpretation of the results.

Analyzing the Alternate Assignment Method

While alternating order based on arrival time may create reasonably comparable experimental groups, it's arguably not as accurate as truly random assignment. There may be confounding variables, such as early arrivers being hungrier or more eager to eat than later arrivers. This could potentially skew the results.

Analyzing the Incorporation of Blinding

The researchers chose to incorporate this type of blinding to reduce bias. If the assistant knew which group the plate belonged to or the purpose of the experiment, it might inadvertently influence the results. For example, they might unconsciously overestimate calorie counts for the large bowl group if they knew the hypothesis.

Key concepts

Random Assignment in Research

Random assignment is a cornerstone concept in experimental design that helps to ensure the scientific validity of the findings. When participants are randomly assigned to different groups, such as control and experimental groups, researchers can be more confident that the groups are comparable in key aspects at the start of the study. This comparability is crucial because it helps to mitigate the effects of confounding variables—those pesky extraneous factors that could otherwise affect the outcome of the experiment.

In the context of the Super Bowl party experiment, participants were assigned to different buffet tables based on their arrival time. If this method truly mimicked random assignment, each group would have a mix of all types of participants, making arrival time irrelevant in influencing the amount of food served. However, it might be the case that certain types of people tend to arrive earlier or later, which could introduce unwanted variables into the experiment. For instance, early arrivers could potentially be more health-conscious and thus serve themselves less, or they could be hungrier and serve more. Such variables would need to be acknowledged and controlled for to draw reliable conclusions about the impact of bowl size on food consumption.

Blinding in Experiments

Blinding is another essential technique employed in research to prevent bias. It involves concealing certain information from participants, data collectors, or researchers to prevent it from influencing both the behavior of the subjects and the interpretation of the data. In the serving bowl experiment, the research assistant was unaware of both the hypothesis being tested and the group membership of the plates when weighing them. This meant they could not subconsciously let their expectations about the study’s purpose or their knowledge of which group a plate came from affect their measurements.

Implementing such blinding helps maintain objectivity in the data collection phase. Without blinding, psychological biases could come into play, subtly altering how measurements are taken or interpreted. For example, if the assistant expected individuals with larger bowls to serve more food, they might unconsciously exert a slight pressure on the scale when weighing plates from the large bowl group, thus tainting the results.

Confounding Variables

Confounding variables, or confounders, are factors other than the independent variable that might affect the dependent variable in an experiment. These variables can create false correlations and obscure the true relationship between the variables of interest. In the Super Bowl study, confounding variables might include the participants' initial hunger levels, their attitudes towards wasting food, or their personal preference for certain snacks. If not appropriately accounted for, these confounders could lead to incorrect conclusions about the effect of bowl size on food consumption.

Researchers must identify potential confounders prior to conducting the experiment and use strategies like random assignment and controlled variables to minimize their impact. However, since some confounding variables might be unrecognized or difficult to control, scientists often use statistical methods to adjust for their potential effects when analyzing data.

Scientific Validity

Scientific validity refers to the degree to which a study accurately reflects or assesses the specific concept that the researcher is attempting to measure. An experiment's validity determines whether the results can be accepted as truthful and reliable. The bowl size study aims to test whether bowl size affects food consumption. For the study to have high validity, the measurements of food taken from the bowls must reflect the actual eating behavior of participants. The design of the study, including how participants are assigned to groups and how bias is controlled for, affects the study's validity.

To enhance scientific validity, the experiment should be replicable with consistent results and free from bias, both in the collection and interpretation of data. To that end, the researchers blinded the research assistant to the details of the study to avoid measurement bias. However, the alternate assignment of subjects based on arrival time may have introduced variability that compromises the study's validity. Additionally, aspects like precise measurements of food and calibration of scales also contribute to ensuring the scientific soundness of the findings.