Our academic experts are ready and waiting to assist with any writing project you may have. From simple essay plans, through to full dissertations, you can guarantee we have a service perfectly matched to your needs.
GET A 40% DISCOUNT ON YOU FIRST ORDER
You have completed the background review of the literature in your literature review, so now your research proposal should seek to design a study to answer the research question posed in the literature review (you can tweak somewhat from your initial research question, but should be in the same general area). You can pretend that you have significant resources (i.e., you got a big grant and can pay for a lots of things!). The research design can be a correlational study, or a true experiment. However, it should remain within the realm of something you might realistically be able to pull off as a researcher (i.e., you’re not random sampling from the entire world and flying people internationally into your lab to complete a 30 minute controlled study; you might arguably say you could try to randomly sample from families in Richmond with a big enough incentive (payment) and bring people in to do a controlled study; you could probably randomly sample from UR college students easily enough). In general, if you are unsure about something, if you think “How would I pay for that,” it’s probably ok for this unlimited-resource proposal; if you think “How would I get people to agree to do that,” you should probably question whether it’s really feasible. Remember people often do very labor-intensive things like spend two weeks living in a sleep lab for enough pay, so there are tradeoffs there. But, you’ll probably have a limitation of some self-selection for those who do not have other commitments they can’t be away from for two weeks.
The proposed research study must take into account principles of ethics, validity, and reliability discussed throughout the semester. If you are in doubt about whether your ideas are appropriate, check with me before completing the paper.
Ethics: The proposed study design must be ethical. Consider principles of respect for persons, justice, and beneficence as you develop your study and write it up – how would you incentivize people to participate (in a non-coercive way – don’t pay people a million dollars to take a survey, no one will say no, but it may be ok to pay $3000 for a two-week live-in study)? Are there any special considerations (e.g., special populations) you would need to take into account regarding informed consent? Are you ensuring you are not recruiting from a population that is bearing an undue burden of risk to benefit others long-term (justice issues)? Do the benefits of your study outweigh the potential harms? What might the potential harms and risks be (this includes things like breach of confidentiality, etc – e.g., if you get mental health information, there is a risk that you can be hacked or lose data and people’s confidential information is released!)
Validity and reliability:
Construct – Are your constructs measured well? Are you considering using surveys that have been validated before? Have they demonstrated good reliability and validity in studies when they were developed (this will often be included in the initial paper from which the survey is obtained). If you are creating your own survey, what types of reliability or validity would you potentially want to assess, and could you do that through your study? If not, think about how that might impact limitations.
Internal – Are you proposing that you are answering a causal question? If so, is your study designed appropriately to do so? Are you thinking about and trying to implement controls to prevent issues with threats to internal validity from chp. 11?
While not internal validity, as part of the causal criteria, you can also think about whether you can establish temporal precedence based on your design. Most designs, by nature, are going to be asking questions about covariance.
External – How are you sampling? To whom is your study going to generalize, if anyone?
Statistical – are you proposing using the right tests? What sample size would you be looking to obtain (this may differ depending on sample size). Think of central limit theorem – how many people might you need for the base assumption that the sampling distribution of the means is normal? Are you thinking about ways you could potentially increase power to detect differences?
Interplay between validities – remember, sometimes these are in tradeoff with each other! So you do not have to optimize each one – but if not, you do have to note that in your limitations section.
Format and Outline
The paper should use APA format, including a title page and a reference page with any references you use in your methodology or discussion section. You will include an additional introduction paragraph as though it continues from your literature review, now with the specific hypothesis(es) you are testing. From there, you will include a detailed proposed Method section, a brief Results section with a proposed data analytic plan, and a Discussion section noting strengths and limitations of your design, and possible implications of your proposed study (meaning, what significance does your study have? Why is it important?). I have given detailed examples below of the types of things to include in each section, but as every study is different, this may not be comprehensive for all studies, so include it if you think it’s important!
The length of this paper will likely vary significantly from person to person depending on the proposed study design and how elaborate it is, so there is no particular page limit or restriction. It is possible this paper could be successfully completed in as few as 3 pages (excluding title page/reference page) with more straightforward designs and measures. However, I expect it is more likely to be in the 4-6 range (excluding title page/reference page), and some might be a little longer.
APA-style title page
Reiterate your research question from the literature review. From that, develop and include the specific hypothesis(es) that you want to test. One or more is fine – if you are proposing an experiment, you might consider only one looking at group differences; if you are proposing correlational research, you might consider looking at a basic association or multivariate association with additional predictors, but could propose testing both (basic association, then is it impacted by additional predictors). Think about whether you would be tracking over time (if so, may want to do a cross-lag panel model); you could also think about possibly suggesting a mediation hypothesis to show a theorized causal sequence. If you are proposing a factorial design, you should be thinking about whether you expect main effects and an interaction, and be clear what you think you would vs. wouldn’t see among those possible hypotheses.
The Method section typically consists of sections such as Participants, Design and/or Procedure, and Materials or Measures (Materials header often refers to physiological instruments like using scales or blood pressure, Measures to surveys or observational codes). The order of these sections is flexible, though Participants generally comes first.
Who is your target population? Be specific – can this be done in a Psyc 100 sample? Would you need to recruit from the community? If so, who are you targeting? Are you recruiting individuals or a unit of some kind (e.g., parent and child, romantic couple)? What is your proposed method of sampling (be realistic – a simple random sample may be feasible for a broad study, but maybe not for finding a rare group)? It’s ok to use biased sampling methods here if it is more realistic, you will just need to note what limitations will arise from that in your discussion section. Are there any specific inclusion criteria (e.g., only people with diagnosed anxiety?).
Think about getting informed consent here – might there be additional levels of consent if you have special populations like children? Will you do anything to protect confidentiality or anonymity? Do you need to for your study? How would you incentivize people to participate?
What demographic breakdowns (gender, age, race/ethnicity, SES, etc) would you hope to obtain? Is this important enough to you that you would consider stratifying in your sampling method? Are you targeting specific subgroups only (e.g., body image in women only, parenting in those below the poverty line, etc)?
Design and/or Procedure
Design will often get used to describe experimental manipulations, including delineating what different levels of the IV will experience. In a correlational study, that section might get dropped and you just write about the procedure for measuring the data. If you have both sections, design will typically outline the basics of what type of design it is, what the groups/manipulations are, and how participants are assigned to them (if between-groups), then procedures will read like a walk-through of exactly what participants would experience when they show up at the lab. You can also combine this into one. In the rubric, this is all assessed together as the presentation of what’s actually happening procedurally to run the experiment, or collect the correlational data.
How are you going to obtain the data? What would a research visit from your team look like from start to finish? Would you go to people’s homes? Bring them into a lab? Send a survey out over the internet? What is the time frame (one study visit, longitudinal over time, multiple study visits? If longitudinal or multiple, what is the time frame between assessment points?)
If you are conducting an experiment, what would the lab setup and manipulation look like? How will you assign participants to groups (if between-subjects)? Are there any possible third variables you might want to consider using for matching purposes?
You can pretend you have research assistants – who will give participants the surveys vs. run manipulations, etc – same person, different people? Will any of those people need to be blind to study hypotheses? Do you need confederates? What will each level look like (in terms of the manipulation)? What design specifically will you use – between-groups? Within-groups? Which specific variant (pretest-posttest? Repeated measures? Etc)? If you are thinking about something like an intervention study, how many sessions would your intervention be? What, broadly, would you be hoping to teach in your intervention?
Are you planning to use deception? If so, what would you tell participants to mask the intent, and what would you tell them in debriefing?
Materials and/or Measures
What will you use to collect the data? If using self-report, what surveys will you use (find ones that exist and cite, or, create your own questions!)? If you use a previously existing one, what reliability information is already available (e.g., has it been shown to have internal reliability in other studies? Test-retest?).
If you create your own, what reliabilities or validities would you need to demonstrate? If you plan on using observational coders, what would you need to demonstrate? If you are doing an experiment, are you using a paradigm (manipulation) that has been used previously (if so, cite it)? If you are creating your own, do you think there would be any benefit in considering pilot tests or manipulation checks to ensure that your experimental paradigm is “working”? If so, what could you do?
Describe each measure. What information is it obtaining? How is it scaled? Does it need to be averaged? What do higher and lower scores mean? Who is completing it – is it self-report, rated by an outside party, etc? Strongly consider using PsycInfo here to identify measures that already exist that you can cite – think about what people used as measures in your lit review! What is a sample item, if the measure already exists? If you create your own, what questions will you be asking? You can reference an appendix at the end and put the questions there (that is, write the questions on a page that looks like a questionnaire, instead of trying to write them all out in a paragraph).
Are there other variables you would want to collect as control variables (e.g., demographics?)? If so, what would you ask or how would you measure?
As we do not have actual data, this section will be brief. Here, include a proposed data analysis that outlines how you would analyze your data. Reference your hypothesis again, then describe what statistical test(s) you would need to run to answer the question. Be clear about what would be the IV(s) and DV(s) here. What results would support your hypothesis? (If your hypothesis were supported, what would you expect to significant in your analysis? What direction of association or what group(s) would be higher or lower?)
If you create your own measures, would you need to run any initial tests to establish construct reliability or validity (think chapter 5)?
Here, you will focus on the strengths and limitations of your design, as well as possible implications of your study. This section should include 1-2 paragraphs each for strengths, limitations, and implications. It is fine to be concise, but be thorough – show me that you recognize what your study is doing well and what potential issues there are with your design, and that you understand why those are good things or problems. No design is perfect – so it is fine to use a correlational study, or a biased sampling method, but if you do, you must then tell me here how that limits the interpretation of the findings.
Strengths and limitations
What does your study do well? Think about this in terms of the big 4 validities (internal, external, construct, statistical). Do you think you set it up in a way that allows for a causal statement? Does it allow for good generalizability? Is your measurement solid? Do you have what you’d need to get statistical significance (sample size, elimination of noise/unsystematic variability, etc)? Potential things to consider here (including but not limited to) are strong sampling methods, good use of random assignment or repeated measures designs with appropriate controls for threats to internal validity, being able to establish temporal precedence, strong measures that have been previously well validated(good construct validity – chp 5 & 6!), good sample size for power, etc! Why are these strengths and how do they help strengthen the conclusions of your study?
What are the limitations of your study? Again, think about this in terms of the validities. Are there threats to internal validity that you think could be present in your design? Are there third variables that could be playing a role in correlational studies that you didn’t or couldn’t take into account? Is your generalizability limited? Are there possible flaws with your construct validity (both in terms of design of the manipulation, or issues with the measurement of variables if you used survey or observational research)? Any concerns about power or statistical validity (sample size? Possible issues with individual differences, or did you do enough to equalize those?). Why are these limitations and how do they limit the interpretation of your findings, the applicability of your findings, etc?
Overall, this section should show me that you understand what good design elements of a study are and how they contribute to strengthening the impact of the findings, and why potential problems in the study design limit the interpretation of or the impact of the findings. It is fine to have an imperfect study design with limitations – just be clear here to identify what they are and show that you understand how they impact the findings, especially in the area of the four big validities. You will only lose points for flaws in your study design if you don’t acknowledge them or don’t show a full understanding of why they are flaws and what impacts the might have.
This will tie somewhat back into the literature review, but here, you will think about why is this study important. How is it adding to the literature? What is it telling us that we don’t already know? This could include (but is not limited to!) such considerations as how it might impact policy, be a starting point for new therapies or interventions, provide understanding of phenomena in a new group, contribute to understanding limitations of how far a phenomenon may go (moderation!), provide further evidence for a causal statement (e.g., mediation and answering whys, experiments when most work has been correlational), or contribute as a springboard for future research studies. You may find yourself citing your background articles again here if it makes sense.
Include any references that you cite in the current paper – this may be in methodology (citing surveys you find, or experiments you are borrowing from) or in the discussion (you may choose to use citations to support statements in your discussion such as implications).