The analysis of covariance (ANCOVA) is a statistical method in research studies utilized in determining the effect of the independent variable on the dependent variable while eliminating the impact of covariate factors. This technique measures the effect of up to 10 covariate factors while using the statistical package of social sciences (SPSS) software. It involves the use of covariates as a control group in the analysis. The presence of covariates necessitates its use over the analysis of variance (ANOVA) method (Armstrong & Henson, 2002). This method also requires the upholding of several assumptions to increase the validity of its results. First, the independent variable may be categorical unless one is covariate, while the dependent variable must be covariate. Second, each group formed by the categorical covariates should have uniformity of the covariate regression coefficients (Armstrong & Henson, 2002). The lack of adherence to this principle increases the risk of type one errors.
The authors in this article by Armstrong & Henson (2002) chose the ANCOVA because of several reasons. First, the ANCOVA is the most widely used method in counselling research. This is due to the use of intact groups with small sample sizes. Second, numerous researchers have often confused the ANCOVA strategy as a methodological design as opposed to statistical analysis. They believe that by using this strategy, their methodological flaws would somehow disappear. Therefore, these deficiencies in practice prompted the critics to come up with this article to address the sources of poor technique (Armstrong & Henson, 2002). The critics aimed to identify the best ways to practice using the ANCOVA.
The ANCOVA test was the most appropriate for this article. This is because it allows for the determination of the effects of the other unobserved factors. The impact of the covariant on the outcome can be measured while using this process of statistics. The power of the study is increased by analyzing the effects of the covariates. Participants are assigned to treatment and control teams to reduce the impact of the unwanted characters, thereby increasing the sensitivity of the findings. Moreover, inserting a stronger covariate increases the power of the ANCOVA. Furthermore, proper utilization of this technique reduces the incidence of type 2 errors in research (Armstrong & Henson, 2002). This is because it is easy to calculate the relationship between the covariate variable and the grouping variable by calculating the Pearson’s r between the variables.
The researchers did not utilize a table in presenting the study’s findings. Instead, the scholars indicated a table need to be inserted there. Furthermore, the authors did not utilize figures to offer data. The inability to use tables or figures to present information reduces the visual appeal of the presented data. It is not easy to deduce the research findings at a single glance. However, the authors did explain the constituents of the tables. The results of the tables do not stand alone, as the authors provide further explanations of the findings. For instance, in table 2, the articles analyzed that verified the uniformity of groups used statistical significance to measure whether the slopes were parallel. This approach is restricted by the sample size that the researcher’s employs in the study (Armstrong & Henson, 2002). Moreover, table 2 indicates that many articles analyzed failed to uphold the test for homogeneity and uniformity. Therefore, researchers need to effectively report on the sample size and consider the assumptions for the validity of the ANCOVA findings.
Reference
Armstrong, S. A., & Henson, R. K. (2002). Current use (and misuse) of ANCOVA in counseling research. Web.