DISCUSSION THREAD: SELECTING AND INTERPRETING INFERENTIAL STATISTIC 2
When evaluating a variety of user interfaces in a single study, researchers may divide the
test subjects into two categories. The initial technique is the between-subjects or between-groups
research design. Therefore, it is imperative to assess each situation with a distinct individual to
ensure that each individual is only exposed to a single user interface. The subsequent design is
the repeated-measures, or within-subjects, study. The same individual conducts testing on each
scenario, which encompasses each user interface. I will assume that the objective is to compare
two distinct vehicle rental agencies, A and B, by examining the process by which clients reserve
vehicles at each service, for the sake of argument. The investigation could be organized using
one of two methodologies. As part of the between-subjects design, each participant in the
research was assigned the responsibility of evaluating and reserving an automobile through an
isolated online platform. In contrast, a within-subjects design guarantees that all participants
have the opportunity to visit both car rental locations and make reservations at their own
discretion. Ultimately, researchers must determine whether to employ a between-subjects or
within-subjects design when conducting user research with a diverse array of test conditions.
Nevertheless, this distinction is essential for quantitative research (Morgan et al., 2020).
D4.5.2. What information about variables, levels, and design should you keep in mind
to choose an appropriate statistic?
Prior to conducting any research, determine the number of variables that each topic or
hypothesis contains. Basic statistical analyses are adequate for scenarios that involve two
variables. Attention should be given to the measurement level of the dependent variable.
Conduct a test using independent samples to determine whether the measurement is on a normal
scale and whether the assumptions are not violated (assuming the independent variables have
two levels). Use partnered samples and verify for repeated readings. When there are more than
DISCUSSION THREAD: SELECTING AND INTERPRETING INFERENTIAL STATISTIC 3
two groups in a between-group design or when there are three or more groups for the
independent variable, a one-way ANOVA is used. Substitute the WILCOXON test for the
independent sample t-test when the dependent variable is evaluated on an ordinal scale, the
KRUSKAL-WALLIS test for the one-way ANOVA, and the Friedman test for the GLM
repeated measure ANOVA. For binary or categorical dependent variables, you can use the
independent samples t test with a chi-square or MCNEMAR test, the one-way ANOVA with a
chi-square or MCNEMAR test, or the repeated measures ANOVA with a COCHRAN Q test.
When dealing with three or more variables, it is imperative to employ sophisticated statistical
methodologies. Using the general linear model, a multivariate analysis of variance (MANOVA)
may be conducted when assessing two or more dependent variables that are somewhat
interrelated (Morgan et al., 2020).
D4.5.3. Provide an example of a study, including the variables, level of measurement,
and hypotheses, for which a researcher could appropriately choose two different statistics
to examine the relations between the same variables. Explain your answer.
Hypothesis: In the context of research or examinations, students who are distracted are
more likely to make errors. IV: Distraction levels were arranged in descending order of intensity:
dining, television viewing, messaging, phone conversations, and socializing with friends. The
total number of DV errors. To compare the levels of distraction among different groups that
interact with children, it may be advantageous to implement an ANOVA. In this context,
conversing on the phone may be classified as a level one distraction, while sending and receiving
text messages is classified as a level five diversion. Morgan et al. (2020) propose that correlation
analysis may be implemented to determine whether there is a correlation between errors and
detours, as there is always a possibility of producing an error.
DISCUSSION THREAD: SELECTING AND INTERPRETING INFERENTIAL STATISTIC 4
D4.5.6. What statistic would you use if you wanted to see if there was a difference
between three ethnic groups on math achievement? Why?
The best method for determining whether there is a discrepancy in arithmetic
performance among three ethnic groups is the Analysis of Variance (ANOVA) method. Data that
pertains to a single quantitative dependent variable and a single categorical independent variable
may be analyzed using a one-way ANOVA. There should be a minimum of three levels in the
independent variable, which translates to a minimum of three distinct groups or categories. In
this investigation, Math Achievement serves as the quantitative dependent variable, while ethnic
groupings, which include three levels, serve as the categorical independent variable. To evaluate
statistical significance, the F-Statistic is implemented in ANOVA. According to Morgan et al.
(2020), the ANOVA table, which enables the simultaneous comparison of multiple means,
produces the F-statistic.
D4.5.8. What statistic would you use if you had one independent variable, geographic
location (North, South, East, West), and one dependent variable (satisfaction with living
environment, Yes or No)?
The initial step is to determine the number of variables that are included in the
investigation. Performing a rudimentary statistic is recommended when there are only two
variables. Given the existence of two variables, one of which is independent (geographic
location) and the other dependent (satisfaction with the living environment, either yes or no), the
fundamental statistical principle is applicable. The dependent variable's measurement level
should be observed thereafter. For nominal or dichotomous dependent variables, the Chi-square
test statistic is the appropriate statistical test to administer. The Chi-square test statistic formula
DISCUSSION THREAD: SELECTING AND INTERPRETING INFERENTIAL STATISTIC 5
is valid because the dependent variable can only be reported as either "yes" or "no" (Morgan et
al., 2020).
D4.5.9. What statistic would you use if you had three normally distributed (scale)
independent variables (weight of participants, age of participants, and height of
participants), plus one dichotomous independent variable (academic track) and one
dependent variable (positive self-image), which is normally distributed?
The most suitable statistical analysis for this investigation would be multiple regression.
The analysis is suitable if there are three independent variables (participant weight, age, and
height) that are normally distributed on a scale, one independent variable (academic course) that
is dichotomous, and one dependent variable (positive self-image) that is normally distributed. To
evaluate the correlation between a single dependent variable and multiple independent variables,
multiple regression is implemented as a statistical method. Utilizing established independent
variables, multiple regression analysis is principally intended to forecast the value of a single
dependent variable (Morgan et al., 2020).
References
