Review of Basic Statistics: The Effect of Color on Emotions Essay

Description of Research Interest

As a determinant of perception through sight, color has a marked and varied influence on the emotions of individuals. Businesses and marketers need to capitalize on colors and color schemes in eliciting a positive emotional impression among customers. In this view, the area of research interest is to determine the effect of color associated with gift wrappers on the emotions of potential customers.

List of Variables and Type of Measurement

1. Age of customers (Ratio scale)
2. The color of gift wrappers (Nominal scale)
3. Emotional impression (Nominal scale)
4. Emotional rating (Interval scale)

Hypotheses

• H₀: The color of gift wrappers and customers’ emotional impression do not have a statistically significant association.
• H₁: The color of gift wrappers and customers’ emotional impression do have a statistically significant association.

Research Approach

The analysis of the research design indicates that the experimental approach is appropriate for the study of the research area of interest. Since the study seeks to control how and where participants view their gifts, researchers need to manipulate variables and the research environment, hence, making the experimental approach appropriate. In essence, researchers need to reveal the effects of color but eliminate the influence of confounders, such as light, gender, race, and biases. Moreover, the researcher assumes the existence of a cause-effect relationship between the variables of interest (Trochim & Arora, 2015). In this case, the color of gift wrappers is the independent variable that influences the emotional impression of customers, which is the dependent variable.

Cost-Benefit Analysis of Research Approaches

One benefit of the experimental approach is that it allows for the determination of the cause-effect relationship between color and an emotional impression. Another benefit is that the experimental approach enables researchers to control variables by eliminating confounders, such as light and age, and introducing the color of gift wrappers and an emotional impression as the variables of interest. Nevertheless, the experimental approach is prone to human errors, researcher biases, and low validity due to the manipulation of colors, color schemes, and the environment where the experiment occurs.

A chief benefit of the correlational approach is that it permits researchers to establish the direction and magnitude of the relationship between colors and an emotional impression. An additional benefit of the correlational approach is that researchers can collect more and cheaper data than would be the case in the experimental study, which requires the creation of a laboratory and the provision of appropriate materials (Creswell, 2014). However, setbacks of the correlational study are that it does not allow the determination of the cause-effect relationship between colors and emotional impressions, and it only applies when color and emotional impressions are on a continuous scale.

Reliability versus Validity

Both reliability and liability are integral aspects of the study for determining the influence of the color of gift wrappers on an emotional impression of customers. The assessment of emotional impressions reveals that they are susceptible to demographic variables, experimental conditions, and researchers. To generate constant and consistent outcomes in emotional impressions, the study should measure up test-retest, inter-rater, and internal consistency forms of reliability ratings. Furthermore, the study ought to provide valid outcomes because the existence of extraneous variables in experimental conditions reduces validity. Questions that measure an emotional impression and emotional rating should meet face validity, content validity, and construct validity to attain outcomes with appreciable internal validity. A sampling method should ensure the representation of the target population of customers to provide findings with substantial external validity (Creswell, 2014). Since reliability and validity have an intricate link and determine research outcomes, it is impossible to apply one of them without the other. Essentially, for a test employed in a study to be valid, it must be reliable.

Sample versus Population

A sample is a manageable subset of the population that researchers select and allow their participation in the study. In contrast, a population comprises the total number of individuals from which researchers select a sample. In this case, the customers selected to participate in the study represent the sample, whereas all customers in the target market form the population. The understanding of the difference between the sample and population helps researchers to make rigorous inferences during extrapolation and the application of the findings. For instance, the nature of customers selected and interviewed regarding the effect of color on their emotional impression should reflect patterns and trends in the population.

Measures of Central Tendency

The following table (Table 1) offers measures of central tendency, namely, mean, mode, and median, of individuals’ weight (N = 20). The distribution of data exhibits a positive skew, that is, the mode and median are less than the mean. In such cases where a skew distribution of data occurs, the median is a statistic that provides the best description of the data. The presence of outliers in data, such as 250 and 275, caused the distribution to skew and distort the mean as a measure of central tendency.

Table 1.

Measures of dispersion point out that the distribution of data tends to spread with a range of 175 and an interquartile range of 12. The standard deviation and variance of 47.165 and 2224.513 respectively show that the distribution of data varies significantly from the mean. The interquartile range of 12 demonstrates that most data points are within the first quartile and the second quartile, and the outliers contribute considerably to the apparent spread of data.

Table 2.

Statistical Significance

If the study had conducted a chi-square test to determine the nature of the association and established the presence of a statistically significant result, it would cause the null hypothesis to be rejected. The rejection of the null hypothesis would mean that the alternative hypothesis is that the color of the wrapper and customers’ emotional impression of gifts have a statistically significant association. Effect size depicts the degree of the influence of color on the emotional impression of customers. From a practical standpoint, it would imply that the color of the gift wrapper influences the emotional impression that customers have of gifts they give or receive from their friends. Thus, businesses and individuals need to consider the emotional impression that gift wrappers create among customers and friends.

Type I and Type II Errors

In inferential statistics, type I and type II errors are two types of inaccuracies that are likely to occur. Type I error occurs when the hypothesis test causes the rejection of the right null hypothesis. According to Trochim and Arora (2015), the occurrence of type I error leads to false-positive outcomes in a study. In contrast, type II error happens when the hypothesis test does not reject a false null hypothesis. As an implication, type II error results in false-negative conclusions in inferential statistics.

Areas of Concern

Areas of concern comprise the design of the questionnaire and Likert scales to enhance the collection and analysis of appropriate data. In this view, how do researchers ensure that their instruments meet particular aspects of reliability and validity?

References

Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods approaches (6th ed.). Thousand Oaks, CA: SAGE Publisher.

Trochim, D., & Arora, K. (2015). Research methods: The essential knowledge base (2nd ed.). New York, NY: Cengage Learning.