Psych 2300: Research in Psychology [Ch 3: Validities & Ch 4 start]
Psych 2300: Research in Psychology [Ch 3: Validities & Ch 4 start] Psych 2300
Popular in Research Methods In Psychology
Popular in Psychlogy
This 12 page Class Notes was uploaded by MadsSwart on Monday June 20, 2016. The Class Notes belongs to Psych 2300 at Ohio State University taught by Emanuele Rizzi in Summer 2016. Since its upload, it has received 77 views. For similar materials see Research Methods In Psychology in Psychlogy at Ohio State University.
Reviews for Psych 2300: Research in Psychology [Ch 3: Validities & Ch 4 start]
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 06/20/16
Psych 2300: Psychology Research Methods 6/15 & 6/17 Four Validities - Validity: appropriateness of a conclusion or decision Construct Validity Operationalization of a variable Applies to all claims Is this a reasonable way of quantifying what I want to understand Assessing creativity measure x : is x a reasonable way to measure creativity External Validity Generalizability o How was the sample gathered o Can I apply this elsewhere Can these results extend beyond my study Can I apply it to the population I am interested in Who can I apply these results to Representativeness How representative is your sample of the population of interest Specifically interested in this for FREQUENCY CLAIMS Are psych 101 students representative of: o College students? Yes. o General population? Somewhat. o Ceo’s? No. - Are business owners representative of: o College students? No. o General population? Somewhat o Ceo’s. Yes. Statistical Validity - How well did you measure the numbers; stats - Did this capture the true result or is there an error in the result found - Extent statistical conclusions are accurate/reasonable Margin of error- how much error are there in my data How different are the two groups if I am trying to claim diff/similarity - We know that the true happiness score is somewhere in this region but we are unsure by this much. - With different margin of errors for each state, you can’t compare them because the positivity you had for each state was different o May not matter for happiness, but would matter for pill effectiveness; death probability etc. Internal Validity Was I able to remove confounds Was there anything else that could have caused it Specific to each claim… - Association claims Psych 2300: Psychology Research Methods 6/15 & 6/17 o Construct (variable operationalization) and external validity (generalizability) still apply o Study two or more variables relatedness Correlational study o Statistical Strength of the relationship Quantified by the correlational measure r Was the relationship by chance Statistical significance Did your study just happen by chance or how statistically valid is it; how representative is it of what is ACTUALLY going on in the environment - Statistical significance o Sample; draw a correlation; is it accurate o Avoid type one and type 2 error True relationship between variables True relationship between variables What you found Correct Type 1 false + as relationship Type 2 false - correct between them - Causal claims - Construct; variable operationalization - External; generalizability - Statistical; relationship strength; accurate conclusions o Check criteria for causation: Covariance temporal precedence internal validity - Control variables: EXPERIMENTATION o What makes it an experiment? An independent (manipulated) and dependent variable (measured) Negating causation: looking to disprove Positive causation: establish a relationship/cause - Internal validity o Third variable criterion Is there some alternate explanation for results o Control for confounds o RANDOM ASSIGNMENT Manages internal validity by making groups in your study similar Make comparison groups as similar as possible to make sure there is no other reason for the result to happen other than the independent variable, as that should ideally be the only thing that is different between the groups - Music lessons enhance IQ: took children and told them to take either keyboard, voice, drama, or no lessons, then measured their increase in IQ scale o temporal precedence: measured after Psych 2300: Psychology Research Methods 6/15 & 6/17 o Covariance: drama & no lessons kids increased, but the music kids increased more o Internal validity: have a comparison group: measured IQ before so would have a comparison group; a number to compare the increase to Have four groups Is it music class or that they are in a class Music: keyboard or voice Drama/no lessons is non music Drama shows that it’s not that just extracurricular or an extra class would increase IQ o Confounds: Random assignment important? Other things that could affect IQ Preference: smarter kids choose music classes Children from different school districts Parent involvement o Generalizability They were 6 year olds Results would apply well to kindergarteners, but not adults Prioritizing validity - Different studies have different goals - Not easy to always achieve all four validities - Frequency or associative claims don’t care about - Associative just want to know that there is a relation o Teaching new language leads to fewer Temper tantrums with children External validity: Generalizability to other children everywhere Statistical validity: your findings are right Internal validity: show it’s not something else causing the lessening o Texting and driving: frequency claim Statistical validity: want accurate conclusions: but not important how sure you are; can just make claim External validity: Generalizability is important o Music increase IQ Not necessarily external validity because can worry about generalizability later Internal validity more important: want to be sure something else isn’t increasing IQ Frequency Association Causal claims claims claims Construct How well have you How well have you How well have you validity measured the measured the measured the variable in variables in the variables in the question? association? study? Statistical How large is the How strong is the Is there a Validity margin of error? association? difference Psych 2300: Psychology Research Methods 6/15 & 6/17 Is the association between groups, statistically how large? significant? Is the difference statistically significant? Internal Validity Experiment conducted? Three criteria for causation met? External Validity How How How representative is representative is representative is the sample? the sample? the sample? How generalizable How generalizable is the association?are the results? 3 claim types - Frequency o Prevalence of a single variable in a population - Association o Relatedness [correlation] between 2+ variables - Causal o One variable having an effect on another variable (or that it does not effect it) Four types of validities did they go about the experiment etc. the right way so that their claim can be trusted? if they violated the validities; we know to not take the claim as strongly - External o Generalizability; do your results apply to the population you’re applying it to; how well can you apply them to other populations o How was your sample taken - Internal o Only applies to causal claims o There is nothing else accounting for the relationship that you see in causal relationship o Controlled for all other possible causes of variable 1s change to cause variable 2s change - Construct o About measurement and operationalization of a variable [because of conceptual variables] if there was not a conceptual variable; just how well you measured it - Statistical o How accurate your measurement/results are o Frequency Accuracy in measuring the frequency claim Margin of error Small margin of error means we are close to the true value o Associative Strength of correlation; how strong they are correlated Are they correlated enough that we care? Statistically valid Didn’t make type 1 or type 2 error o The results reflect what’s actually the truth in the environment o Causal How strongly the one variable influences the other With groups, differences between the comparison groups Differences and strengths of results INTERROGATING HEADING: Homework 1 practice Does the heading match the info o Frequency claim o Association Claim o Causal Claim If the heading seems to imply a ____ claim, does the information imply the same claim? Ask a question or two in each validity type for that claim 1. One hour of extra screen time drags down teenagers grades; only followed students around a. claim: associative that looks like causal because of ‘drags down; b. variables: screen time and how engaged/energetic c. Conceptual Variable: d. External Validity Question: students to students is probably fine; but for an overall population of people, would need to be questions e. Internal Validity Question: n/a no temporal precedence or covariance f. Construct Validity Question: they just followed them around; is just having the screen open have an effect or would they must be engaged? Just during school hours? g. Statistical Validity Question: effect but no correlation reported; because you found it in your sample; does that mean it applies to real life h. Was any type of validity prioritized? 2. Most caregivers look after elderly parent; invest a lot of time a. Heading claim: Frequency b. Info claim: interested in how much time was spent as a caregiver; the presence of something [time]. They’re just throwing stats at you! i. Population: employed/professional caregivers ii. Amount of time caregiving to parent iii. Prevalence caring for someone older than 75 iv. Prevalence of taking care of elderly parents c. Conceptual Variable: How compassionate someone is, altruism, empathy, sense of responsibility, etc. could be measured by how much time they spend caregiving d. External Validity Question: population; employed/professional caregivers. Was it well defined? e. Internal Validity Question: N/A f. Construct Validity Question: how they define looking after an elderly parent? i. Giving medication, helping in and out of bed, visiting them _ times a week? ii. Were all participants in survey providing the same type of care? g. Statistical Validity Question: did they accurately estimate caregiving? Margin of error? h. Was any type of validity prioritized? i. Would you have prioritized the same validity? 3. Heavier babies do better in school; measuring birth weight and performance on SAT a. claim: Association i. Weight & Performance scores are related; yet one does not cause the other ii. Conceptual Variable: fetal health to later academic performance b. External Validity Question: What type of school was used to measured, private or public, parent’s smarts and involvement, Extra- curricular activities, Socio economic factors etc. c. Internal Validity Question: N/A d. Construct Validity Question: Is the SAT a good measure of intelligence? Does fetal weight measure their health; weight does not equal health. e. Statistical Validity Question: ¼ of the top 5% of SAT scores weighing 8lb at birth vs 1/3 students weighing 8lb at birth; does this show that heavier babies do better in school? f. Was any type of validity prioritized? i. Construct and statistical 4. Go green; people spend more for environmentally-conscientious products and services: growing number of people are willing to pay for eco-friendly services and products through survey a. claim: Frequency Claim i. can look at a causal claim: spend more ii. population: everyone; general global public; multiple countries iii. Conceptual Variable: product purchasing; how can we change people’s product purchasing habits? iv. Willingness to pay more for eco-friendly products v. Country b. External Validity Question: Are the countries that they measured/surveyed representative of all the countries/countries they did not measure. c. Internal Validity Question: N/A d. Construct Validity Question: How was the survey set up; lickert scale – was the scale a good way to measure one’s willingness? weak i. Asking someone how willing they are to pay for a product is not the same as them actually spending money on a product. e. Statistical Validity Question: How far away is their actual willingness to pay from the willingness to pay that was found/measured through the study? f. Was any type of validity prioritized? 5. Art makes you smart; took a lottery of students at a school and those who won, got to go to the museum. Those who lost did not. They had both sets write an essay and analyzed them a. claim: Causal i. Students who went through the museum; lottery winners ii. Students who did not go through the museum; control b. Population: high school students c. Conceptual Variable: measuring intelligence d. External Validity Question: generalize to other situations i. Different majors of students etc. e. Internal Validity Question: i. Covariance ii. Temporal precedence iii. Didn’t measure control group iv. Lottery is fairly random v. CONFOUNDS 1. Different teachers leading through museum giving different levels of instruction 2. Some kids go through interactive art museum and others through history of art museum 3. Were there differences among school districts or classrooms-more external 4. Same amount of time at the museum 5. Did they see the same exhibits f. Construct Validity Question: returning to the museum was how they measured if intelligence was improved; how does returning to the museum show an improvement in intelligence? i. We are assessing your entire intelligence on ability to write an essay ii. Art engagement; what if some kids don’t like art iii. Essays were about art; kids who didn’t visit museum didn’t have immediate education iv. Does visiting a museum mean they were actually engaged in art appreciation g. Statistical Validity Question: Where was the school? What were their demographics/socioeconomic factors? We need to know more about the school/students i. How strong was the effect ii. What was the difference in scores between museum and no museum group iii. Is the difference reliable/accurate? h. Was any type of validity prioritized? ALL END OF CHAPTER 3 Psychology 2300: Research in Psychology 6/20/16 Chapter 4: Ethical Research - Big 3 principles - Objective 3 History - When you think about human experimentation o The type of idea that we are doing psycho movie type of stuff on people is from history Tuskegee - Had syphilis - Wanted to just follow them around and see the nature of the sickness - Treatment [penicillin] was found, but those in the study weren’t told that because they wanted to keep monitoring them - Started in 30s, out in media in 60s, 70s it was stopped, 90s was when gov apologized - Why many [especially low income and African americans] don’t trust government - Allowed for laws to come into place Milgram - Nazis asked how they could hurt people: just following orders could that be a defense? - Could people obey something even though it goes beyond their moral standings - Shocking in other room: how far up the shock level will he go to shock an imaginary person because they are being told to keep shocking them o Wanted to know how people responded to authority to enact in unjust acts Prison Study - Zimbardo [Zimbardo & Milgram went to same high school in Brooklyn] - Assigned random roles: prisoner or guard - Students took on roles very aggressively o Mistreatment of fellow students o Not allowed to leave - Meant to go on for 3 weeks; his girl friend came in after 6 days and told him to shut it down because of how insane it was Something needed to be put in place to keep unethical studies like this to keep happening Core ethical principles: BELMONT REPORT Provided a guideline for what constitudes all research Psychology 2300: Research in Psychology 6/20/16 - Beneficence o Cost/benefit analysis for participants and society - Respect for persons o Informed concent; are people completely aware of what is going on Can they leave whenever they want o Protection for vulnerable populations Make sure people giving consent know what they are saying ok to - Justice o People taking on risks for study should take on benefits for studies o How are participants selected Belmont Report vs. APA Ethical Principles The Belmont Report is for ALL research. The American Psychological Association’s Ethical Principles is specific for Psychological research http://www.slideshare.net/gemesse/apa-ethics-code I have highlighted all of the main ones in addition to ones we covered in class. Check out the following link for the actual APA Ethical Principles [its only 18 pages, and not horrible to understand]. However, it seems like below is even more than we need to know~ Beneficence Beneficence and No maleficence - Do no harm - Protect welfare and rights of involved [including animals & people affected after study] - Guard against all personal factors leading to misuse of professional standing. - Aware of own physical and mental health’s impact on ability to help those working with Fidelity - Making sure subjects understand what is going on - Being open and clear with everyone - Establish relationships of trust with those working with - informed consent the individual in question or participating must give their consent with language that is reasonably understandable to all people [you must explain it reasonably and understandable and they must be clear that they understand it; know 100% they understand]. o For those legally incapable provide appropriate explanation, consider preferences and best interest, and obtain appropriate permission from legally authorized person. Psychology 2300: Research in Psychology 6/20/16 o If court ordered inform the individual of nature of services, making sure they understand that they are being mandated Responsibility - professional and scientific responsibilities to society and specific community - For subjects, for study, for reporting accurate findings etc. - Ethical compliance of colleagues’ scientific and professional conduct Integrity o promote accuracy, honesty, and truthfulness in psychology o Do not steal, cheat, engage in fraud, subterfuge, or intentional misrepresentation o Keep promises and avoid unwise or unclear commitments. o maximize benefits and minimize harm o correct resulting mistrust or harmful effects caused from any unethical acts o Documentation must be provided during all stages of meetings, to ensure transparency Justice Fairness to all individuals - All people must be able to access and benefit from contributions of psychology o The research pool = the benefited pool o Say you have a new HIV drug, but can’t find a great random sample of people with HIV in America. You know that South Africa has a high number of individuals with HIV, due to their lack of access to resources. If you decide to go and try out the drug on people in South Africa, then the people who benefit from the drug have to be those with HIV in South Africa. [In other words, you couldn’t test the drug in South Africa, find out that it works, and then bring it back to America so that individuals with HIV in America could benefit from it.] If South Africans are researched people, they must benefit from your findings. If you want only Americans to benefit, you must only do research in the US. - Equal quality in processes, procedures, and services conducted - Reasonable judgement and as little bias as possible o Make sure you aren’t the only one making ethical decisions Respect for Persons [do NOT engage in]… Discrimination Sexual [or other] Harassment - Behavior such as solicitation, verbal or nonverbal conduct of sexual or demeaning behavior Psychology 2300: Research in Psychology 6/20/16 Multiple Relationships - you have both a professional and personal relationship with someone [OR a professional relationship has a personal relationship with someone you have a personal relationship with] Conflict of Interest rd - agreeing to provide services to a person through a 3 party, everyone involved must fully understand the nature of this relationship [Autonomy] - People make up their own minds - They decide if they want to be in a study, they decide if and when they can leave - Basic rights of research Privacy, Confidentiality, Disclosures & Intrusions on privacy Less autonomous populations who’s autonomy you must protect - Children: Their parents told them they have to do it for them money - Prisoners: Think sentence will be shortened if they do it; due to authority figures above them - Patients: In desperate situation; when you say you have a new pill for their disease, they may believe they are willing to try anything to rid their life of it, but must make sure they understand any and all risks involved Ethical grey area R H S G DON’T DO GREY K H O IT AREA F O N O GREY AREA DO IT G W LOW HIGH BENEFIT OF DOING STUDY
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'