|
Hypothesis: The null hypothesis needs to be written with words, not symbols. Every research question should have at least one corresponding null hypothesis; however, sometimes more than one is needed. The number of hypotheses should be based upon the number of variables under study and planned analysis. Well-formulated hypotheses are based on the following criteria: (a) the hypothesis stated the expected relationship/differences between variables, (b) the hypothesis is testable, (c) the hypothesis is stated simply and concisely as possible, and (d) the hypothesis is founded in the problem statement and supported by research. Like the research questions, the hypotheses in null form directly influence the statistical procedures used.
|
|
Participants and Setting:
Begin by describing your target population. This may include the description of a large geographical area or a school district from which the sample was drawn. Real names should never be used.
Next, describe your sample. The sample size, the type of sample, and the sampling procedures (e.g., convenience sampling, cluster sampling, etc.) must be explained. In other words, the sample selection procedures (who, what, when, where, how) need to be explained in enough detail for the study to be replicated. Include basic demographic information (number of participants, sample size, age, ethnicity, gender, etc.) described in narrative form. Since this is a proposal, plug in “place holders” (e.g., the sample consisted of 00 males and 00 females…). Quantitative literature citations must be provided for the adequate sample size (e.g. For this study, the number of participants required for an adequate sample size will be 66 students which according to Gall et al. (2007, p. 145) will the required minimum for a medium effect size with statistical power of .7 at the .05 alpha level).
Next, discuss the setting (e.g., specific course, program, online/offline environment, semester-term, and/or treatment/control group testing location, etc.). Real names for people and schools should never be used. Use pseudonyms for descriptors when necessary (e.g. high school #1, biology lab # 2). The setting, especially the treatment setting needs to be described in sufficient details so that the study could be replicated. The setting is often intertwined with the description of the sample.
After you have described the sample and setting, you need to identify and describe each group (e.g. treatment, control, etc.). Remember: correlational studies involve two or more variables and only “one group.” Explain the groups’ formations (e.g., random assignment, naturally occurring groups, etc.) and demographic information (e.g., age, ethnicity, gender, grade level, etc.) for each group. Since this is a proposal, plug in “place holders” (e.g., the treatment group consisted of 00 males and 00 females, etc.). Groups must be explained in enough detail for the study to be replicated.
The recommended length is 200-400 words.
|
|
Instrumentation:
In the instrumentation section, the instrument(s) that are used to measure each variable need to be identified. The instruments may be tests, surveys, questionnaires, or other measurements. Only validated instruments may be used and it is not acceptable to propose to develop an instrument for the purposes of this study.
A description of each instrument, its content, its appropriateness needs to be included. The exact procedures for the development for the instrument (i.e. studies to establish validity and reliability, as well as reliability statistics) must be cited. State other peer reviewed studies where the instrument was used.
Scoring information for the composite and subscales must to be included. For example “… the instrument consisted of 30 questions and used a five-point Likert scale that ranged from Strongly Agree to Strongly Disagree. Responses were as follows: Strongly Agree = 5, Agree = 4, Neutral = 3, Disagree = 2, and Strongly Disagree = 1.” Include scoring information regarding the instrument for example, “… the combined possible score on the ATSF range from 20 to 200 points. A high score of 200 means that the student is… etc.” A brief overview of how the instrument should be administered should be discussed and the approximate time to complete the instrument should be reported.
Recommended length is 200–300 words.
|
|
Procedures:
The procedure section is similar to a “cookbook.” It should contain enough detail that a reasonable person can read your procedure and conduct your study and produce the same results. This includes but is not limited to information about securing IRB approval, eliciting participants for the study, conducting a pilot study, training individuals to implement treatment, administration of the procedures, gathering the data, and recording procedures.
If the study involves the training individuals to implement a treatment and/or administration of the procedures. This should include fidelity measures of the treatment. These should measure the extent to which delivery of an intervention stays true to the original model. For example, if the study calls for the implementation of a school-wide intervention, there should be periodic assessments of teachers in the school to ensure that the intervention is being implemented appropriately. This process should be explained in detail.
A detailed description on how the data will be collected and recorded must be provided. The procedures should be described in a chronological, step-by-step format. Remember; describe the procedures clearly and with enough detail so that the study can be replicated.
The recommended length is 200–300 words.
|
|
Analysis:
In the data analysis section, the type of each statistical analysis technique(s) must be identified (t-test, ANOVA, ANCOVA, etc.). A concise rationale for the type of statistical analysis must also be provided. The rationale needs to be supported via the textbook. The chosen statistical procedures must be consistent with the research questions, hypotheses, and type of data collected (in other words, why is the chosen analysis the most appropriate choice to test the hypothesis?).
For each identified statistical analysis technique ALL assumption tests and how they will be tested must be addressed. For example: “Assumption testing included examining Levene’s test for homogeneity of variances, creating scatterplots to test for linearity, etc...” The alpha level for each statistical analysis technique must also be stated. Lastly, the effect size and the convention used to report it should be explained. Example: Eta square interpreted based on Cohen’s d.
In this section, there must also be an identified statistical procedure for each hypothesis. Thus, each section should be organized according to the research hypotheses.
The recommended length is 100–200 words per hypothesis.
|