The research hypothesis: H1: In comparison with other types of interventions, PECS is the most effective in promoting functional communication for individuals with autism.
The Null hypothesis (Ho) is usually the opposite of the alternative hypothesis H1, it is the hypothesis that the research wants to disprove or nullify (Whittaker, 2012). Therefore, the null usually refers to the common view of something. The null hypothesis is used in statistics to indicate that no variation exists between the stated variables (Gravetter & Wallnau, 2004). For this research, the null hypothesis is as stated below.
Ho: There is no difference between PECS and other types of interventions in promoting functional communication for individuals with autism
The independent variables (IVs) and dependent variables (DVs) are research variables. A research variable is anything that has the variation or can vary. The variables in the experimental investigations are called dependent or independent variables. In the research experiment or study, the researchers aim to look for the possible effect that the independent variables have on the dependent variable (Miles & Shevlin, 2001). Therefore, the independent variable (IV) is the variable that the researcher manipulates or changes, it is assumed to have a direct effect on the dependent variable. While the dependent variable (DV) is the variable that the researcher measures after manipulating or changing the IVs that have been assumed to have a significant effect on the DV.
In this research, the dependent variable (DV) is the Picture Exchange Communication System (PECS) in promoting functional communication.
The research, independent variable (IDVs) includes: Compare PECS with other types of intervention.
The Extend for Vocalization and The success rate for conversational speech.
The non directional hypothesis is also called a two-tailed hypothesis. This means that the research predicts that the various groups or two groups will differ, but it does not provide information about the direction it would take. In this type of hypothesis both the null (Ho) and the alternate (H1) are presented (Gravetter & Wallnau, 2004). Therefore, the non directional alternative hypothesis usually states that the null hypothesis is wrong. Hence, in the nondirectional alternative hypothesis does not provide predictions on whether the parameters would be smaller or larger with regard to what is specified by the null hypothesis.
In the directional hypothesis, the researcher predicts that the various groups or the two groups, one of them will do better than the other one. This is where the r searcher predicts that one of the two groups is likely to be greater than the other group. This is where the null hypothesis is equal or less while the alternative hypothesis is greater which is what the researcher predicts. The directional alternative hypothesis also state that the null hypothesis is wrong, and it also provides specifications on whether the value of the parameter would be less or larger than the null hypothesis (Miles & Shevlin, 2001).
Using the directional hypothesis has some advantages because it increases the possibility of detecting the effect the researcher is interested in. The research hypothesis stated above is directional because the alternate hypothesis states that the null hypothesis is wrong, and it also gives the direction of comparison of other interventions, which is supposed to be greater about the null hypothesis (Whittaker, 2012).
When testing the hypothesis, there are two types of errors that may occur which include the type I and type II errors. The risks that these two errors have is usually inversely related and often determined by the level of significance and the test power (Aberson, 2002). The most important thing for the researcher is to determine the type of error that could have severe consequences for the research before defining their risks (Miles & Shevlin, 2001). This means that there will never be a hypothesis test that is a 100% certainty because it is usually based on probabilities.
The Type I error occurs when the null hypothesis is true, yet it is rejected. Therefore, the chances of making the type I is α (alpha) which is usually the level or significance that the researcher sets for the hypothesis test. Hence, an α of 0.05 shows that the researcher is ready to accept a 5% chance of being wrong when rejecting the hypothesis. Therefore, the research can lower the rest by reducing the value of α. In this research, if the type I error occurs, the hypothesis would be rejected yet it is false. Hence, the researcher would be on track as the hypothesis was supposed to be rejected.
On the other hand, the Type II error usually occurs when the null hypothesis is false, but it is not rejected. The chances or probability for the researcher to make a type II error is β (beta) which relies on the power of the test. Therefore, the type II error risk can be reduced by increasing the power of the test through increasing the sample size (Aberson, 2002). In this research, if the type II error occurs, the false hypothesis would be retained causing confusion to the research. This means that the researcher has to repeat the study using a larger sample size.
The research hypothesis above is an associational question which seeks to compare the result of alternative interventions to the PECS intervention with regard to promoting the functional communication for individuals with autism.
References
Aberson, C. (2002). Interpreting null results: Improving presentation and conclusions with confidence intervals. Journal of Articles in Support of the Null Hypothesis, 1, 36–42.
Gravetter, F.J., & Wallnau, L.B. (2004). Statistics for the behavioral sciences (6th ed.). Belmont, CA: Wadsworth
Miles, J.N.V. & Shevlin, M.E. (2001). Applying regression and correlation: A guide for students and researchers. London: Sage Publications
Whittaker, C. (2012). The speech aversion hypothesis has explanatory power in a minimal Speech Approach to aloof, non-verbal, severe autism. Medical Hypotheses, 78(1), 15- 22