Testing Hypothesis the First Step to Testing

Testing Hypothesis

The first step to testing a research topic is conducting a literature review to determine what has been researched, any gaps, or any inefficiencies in the research studies. The literature review aids in determining appropriate hypothesis, variables that need to be studied, and appropriate research study methods and designs. Depending on the research methodology and designs as well as the goals of the researcher, appropriate confidence intervals can be determined to evaluate the level of confidence in the research findings.

Depending on previous research studies, if an old timer stated, "Why, in my day, kids were much more respectful and didn't cause as much trouble as they do nowadays!," an appropriate alternative hypothesis could be "Children growing up in the 1940's were more respectful and caused less trouble than children in the 2000's." From the alternative hypothesis a null hypothesis of "Children growing up in the 2000's were as respectful and caused the same amount of trouble as children growing up in the 1940's."

From the literature review an appropriate definition of respect can be drawn to serve as an appropriate measurement of participant results. A sample of participants of 70 years or older is appropriate because of the topic years, 1940's and 2000's, involved. A questionnaire with questions that address the variables found in literature as well as gaps and inefficiencies of previous studies can be drawn. The questionnaire should be standardized for all participants with the same wording, order of questions, and definitions of each variable or measurement, written with clear understanding and encouragement of giving accurate answers, unbiased, and complete information. The order of questions should have opening questions, questions that flow in psychological order, variety, and closing that allow for multiple responses of specified answers to stay on track and eliminate any biasness in the questionnaire.

The goal of the confidence interval level is to determine to what degree the evidence is strong enough to believe. Lower levels of confidence produces stronger evidence where higher levels of confidence produce weaker evidence. The confidence interval is also the level of significance, or probability, in drawing a correct conclusion based on the association of the prediction and the outcomes of results. It also determines the level of differences between and among groups. Confidence intervals are related to sample size and power, such as lower levels for small samples with negative findings or higher levels for larger samples.

Confidence intervals are determined based on how the sample size represents the entire population of the study. In situations of insufficient sample size or a high degree of variance in scores that produce high risk of error, lower levels of confidence intervals are used. The confidence interval is calculated for each variable to determine the level of significance. An increase in precision occurs from variance, standard error, decreasing as sample size increases and higher levels of confidence intervals are used. Lower levels of confidence intervals are also used when the research is a comparison of previous studies and being conducted to determine meaningful data from each previous study in a meta-analysis study. It assesses the importance of research findings.