Methods and stats review
What percentage of scores fall above 1.96 z? Explain how you know. How is this related to the rule of thumb p< .05?
2.5% of the scores fall above the z value of +1.96 and 2.5% fall below the z value of -1.96 p < .05 and the z-value of 1.96, you are describing a two-tailed hypothesis (no direction has been indicated). In a one-tailed hypothesis (in which direction has been indicated), we would need to find the z-value that corresponds with the proportion of .05 (or 5% rather than 2.5%), as we are no longer dividing up our alpha into 2 tails. This z-value would be 1.645. Reply
Confounding and noise/nuisance variables are extraneous variables. Differentiate from these two types of extraneous variables and indicate which type of validity threatens each and why.
A confounding variable is described as an element of an experiment responsible for the inability or difficulty to tell if the result was caused by the intervention or the confond according to the United State's Department of Education's Institute of Education Sciences (2018). In every type of scientific study a confound is likely to occur. An example of a design with a confounding variable would be six faculty volunteers at a school agree to teach a new curriculum. Three teachers have at least 20 years of experience with the old teaching materials. Three teachers are new to teaching. The confound of this study is the experience of the teachers with past job experience (United States, 2018). Confounds directly threaten internal validity which validates that the independent or manipulated variable is responsible for differences found of the dependent variable (Walinga & Stangor, 2014). Noise or nuisance variables differ from confounds because they are not variables that occur similarly through the different levels of an experiment or study, and cause an increased chance of errors to be calculated ("Nuisance Variable," 2020). An example of a nuisance variable is if a group is asked to perform a task next to a room where construction is being performed. Some individuals may be distracted by the noise more so than others that creates an amont of error in the data ("Nuisance Variable," 2020). If test scores are affected by individuals not able to focus, statistical validity is threatened because the true data is not given (Walinga & Stangor, 2014).
What is a "hypothetical construct"? What is an "operational definition"? What is the primary function of operational definitions? Provide examples.
A hypothetical construct is a construct that is something someone can not directly observe or measure. One example of this is intelligence, instead of physically seeing intelligence we base intelligence on what we think it can be. We construct the idea of what it is by what we know or feel. And how we do observe hypothetical constructs is by operational definitions. An operational definition is the process or operations by which a variable can be defined and observed. An example would be the operational definition of intelligence. We can include terms that are observable measures, such as someone who can read fast or someone who gets 100% on every test they take. These are things that we can observe intelligence by. And this is the primary function of an operational definition since we can not observe intelligence directly we use other measures to conduct specific observations to define something that is a hypothetical construct.
What is Quasi Independent variable? Provide examples of QIV and their levels.
A quasi-independent variable is a particular personal attribute, behavior, or trait that cannot be manipulated by the researcher (APA Dictionary of Psychology). The particular trait is inseparable from an individual. With a quasi-independent variable, researchers do not have full experimental control and participants will not be randomly assigned to the independent variable. The levels of quasi-independent variables are pre-determined where the researcher starts the experimental design by selecting an independent variable and a dependent variable (Thomas, 2020). For example, a researcher discovered that at a particular high school, some teachers offer an older version of ACT practice while other teachers use a new ACT practice. The researcher hypothesized that a newer ACT practice will lead to higher ACT scores among the high school students than the old one. The researcher will use the pre-existing groups of high school students that participate in the older ACT program versus the high school students that use the new one. The groups were not randomly assigned, so the researcher could account for higher scores with the new program based off the treatment and not other confounding variables. The levels of this experiment are the treatment group that receives the new program and the comparison group that uses the older program. Another example of using a quasi-independent variable is if a researcher hypothesized a new exercise program that helps to relieve anxiety among young adults. The researcher will first identify pre-existing groups of young adults that already uses exercise to cope with anxiety. One group will continue using exercise to help cope with their anxiety while the other pre-determined group of young adults will use the new exercise program hypothesized by the researcher. The quasi-independent variable would be the new exercise program that is used to help relieve anxiety. The different levels of the quasi-independent variable will be the young adults receiving the treatment, which is the new exercise program and the other group that does not use the program will be the comparison group.
What is a Variable? What is an Independent and Dependent variable? Provide examples of an IV with its levels. Provide related examples of DVs that might be affected by the IV(s).
A variable can be defined as a (measurable) condition that changes in an experiment. Variables can also contain levels, or subsets within the condition that can be manipulated in the experiment. There are two main types of variables measured in an experiment: independent and dependent variables. An independent variable is the variable, or condition, that is manipulated or changed by the researcher. The dependent variable is the variable affected by the independent variable, or the 'outcome' of the experiment measured by the researcher. Example of IV levels: For example, to determine whether caffeine consumption influences test scores, a researcher may decide to perform an experiment where they assign students to one of two groups. The first group will be instructed to drink one small cup of coffee, while the second group will be given one small cup of water to drink before the exam. In this experiment, caffeine consumption was the independent variable, which included two levels. These levels were the coffee given to one group before the exam and water given to the other group. The dependent variable in this experiment is the students' performance on the exam or their test scores. The levels of caffeine consumption may influence the students' performance on their exams. Another dependent variable that could be affected by the independent variable is the student's concentration levels. For example, those who were given the caffeine before the exam may or may not be able to focus on their exams as well as those who were given water before the exam. Another dependent variable that might be influenced by caffeine consumption (IV) is the time it takes students to complete (turn in) their exam. Caffeine may influence how long it takes students to take the test.
What is the purpose of an informed consent form? What specific information should be on an informed consent form?
An informed consent form is signed documentation of consensual participation. The form is an ethical process that is written in language that can be easily understood, it should decrease outside influence, and the participant must be given time to consider research and its possible risks (Manti & Licari, 2018). The form is required when research involves people of all backgrounds, and when research uses genetic material or samples and any sensitive information of participants (Manti & Licari, 2018). The information that must be present to make the informed consent form valid are categorized into basic and additional elements. The basic elements include the research description, possible risks or discomforts, benefits, disclosure of alternative procedures, specifics for confidentiality, medical and compensation, contacts, and voluntary attendance (Informed consent, 2014). Additional elements could include descriptions covering unforeseeable risks, termination of participation, any costs to the subject, consequences of withdrawal, information of new findings, number of participants, inclusion for clinical trials, and lastly documentation of informed consent (Informed consent, 2014).
mean, median, mode and frequency
Each of these are measures of central tendency. Mean in APA is presented as M (for a sample). Mean is used to calculate central tendency when you have interval or ratio data and the distribution is normally distributed. Median is used to calculate central tendency if you have interval or ratio data and the distribution is skewed. Additionally, median can be used to calculate central tendency if you have ordinal data with 6 or more options. Mode is the appropriate choice for central tendency if your data are nominal or ordinal (with fewer than 6 categories). Also, let's take another look at frequency. Frequency refers to the number of times something occurs. So, given your example, the frequency of each score would be 1 because each score only occurs one time.
What does "effect size" mean? How does it differ from "variance accounted for"
Effect size is how to measure the difference or strength between two variables being measured. For example, if we were to measure different sales between stores, the greater difference in sales, the greater the effect size would be. Effect size also helps in determining if the difference between two variables is real or if it is small enough that there could be factors that throw off the experiment. The most popular way to test effect size is with Pearson's r. The formula is shown hereVariance accounted for is one way to evaluate effect size. Variance accounted for tells us how much variance in the dependent variable that the independent variable can explain. So, if you have a variance accounted for of .20, that would translate into this: The IV can explain 20% of the variability in the DV. There are other ways of evaluating effect size, such as Pearson's r (like you mention & this is used in correlational research) and Cohen's d (used in experimental research). Note that the formula for explained variance should denote how you calculate it. Explained variance = SSbetween/SStotal, or the ratio of between group variance (or variance in scores due to the effect of the independent variable) to the total amount of variance in scores.
Distinguish between internal and external validity.
External validity refers to the extent to which a study's results can be generalized to the wider population, similar samples, or other settings. Most commonly, you'll hear external validity discussed in terms of generalizing results to the wider population. In order to do so, you must have a representative sample. That is, the sample should have the same characteristics as the wider population from which it was selected. Therefore, the sampling technique that you use is related to external validity. A random/probability sampling technique is going to allow for stronger external validity compared to a non-random/non-probability sampling technique (e.g. convenience sampling). Now let's discuss internal validity. Internal validity, as its name suggests, refers to what's going on WITHIN the study -- not how the study will apply to other individuals or contexts outside of the study. Specifically, internal validity refers to the extent to which a researcher can attribute changes in the dependent variable to the manipulation of the independent variable. This is where experimental control, such as random assignment, comes into play. The better we control for extraneous variables, the stronger our internal validity will be. So, for example, say you were conducting a study on the relationship between exercise and anxiety. In this study, you had an exercise group and a non-exercise group but instead of using random assignment, you put graduate students in one group and retired older adults in the other group. Yikes! Internal validity would be compromised because at the end of the study, you wouldn't be confident that it was the exercise that affected participants' anxiety or if it was the type of participant/their stage of life.
Ways to Quantify Behavior: Frequency, Rate of Response, Duration. Provide an example of each.
Frequency: total number of occurrences Example: Jared left his seat 5 times during the 7th period. Rate: Count the number of times the behavior occurred in the time observed. Divide the count by the length of time the behavior was observed. Example: If Jared raised his hand 20 times in a 10-minute observation, the rate would be 2 hand raises per minute (20 hand raises divided by 10= 2 hand raises per minute). Duration: (I found 2 different ways to record duration) Duration 1: Average duration of behavior: Sum the total durations and divide by the total occurrences. Duration 1 Example: During a 60-minute observation, Lucy had 3 tantrums that lasted 3 minutes, 7 minutes, and then 5 minutes with a total duration of 15 minutes. Average duration: 15 minutes divided by 3 tantrums = Average 5 minutes per tantrum. Duration 2: Percentage of observation with behavior: Sum the total number of min/sec/hrs that the behavior occurred during the observation, divide the sum of the total number of min/sec/hrs of the observation, and multiply by 100. Duration 2 Example: During a 60-minute observation, Lucy had 3 tantrums that lasted 3 minutes, 7 minutes, and then 5 minutes with a total duration of 15 minutes. The % of observation with behavior = 15 minutes divided by 60 minutes = .25 times 100 = tantrums occurred during 25% of the observation.
Describe and distinguish among the following Non-Experimental Designs (give an example): Correlational Quasi-Experimental
Hi Ado! Please note that quasi-experimental designs do not typically use groups of 2, it just depends how many levels there are of your quasi-independent variable. Also note that quasi-experimental designs do not use random assignment like true experimental designs. Quasi-experimental designs investigate pre-existing groups (e.g. smokers & non-smokers) for which it is either not practical or ethical to use random assignment. Correlational this a non-experimental design which deals with relationships between two variables and to see if the two variables correlate with each other. However, with quasi-experimental design this designs deals with the cause-and-effect relationship between independent and dependent variables to see if one variable caused the other variable to have an effect and vice versa. Quasi-experimental has two group when doing the experiment such as a treatment and control group. So, the independent variable is getting manipulated in the treatment group whereas the control group is not having a variable manipulated and it is more of a placebo group. Quasi-experimental uses groups of two usually and does noy use a random assignment. Quasi-experimental and correlational design do correlate in a way such as quasi-experiments use correlational study design results to form thesis to test their experiment. Examples of correlational design would be for instance we want to know if students who spend time studying get better grades than those students who do not spend time studying? Examples of Quasi-experimental: You have two classes, and you want to see if the online modules will help increase the student's grades. So, one class will use the online modules while the other class will use the old school teaching method.
What does IRB stand for? Describe its major functions.
IRB stands for Institutional Review Boards. This is required by law, because it protects human participants from harm and makes sure the benefits involved outweigh any risks. Any time human participants are used, APA guidelines as well as the Title 45 of the Code of Federal Regulations states, IRB is required to approve before a study or experiment takes place. When considering what the IRB labels benefits and risks, benefits should be those that improve ones mental or physical state, or can benefit the present or future of man kind. Risks include any physical or mental harm, environmental harm, or violation of human rights which does include privacy.
Pearson's Correlation: What is it? What is the symbol. What does it show? What does it range from, etc.
In statistics, a correlation is a positive or negative relationship between two variables. A positive relationship of two variables means that the variables change in the same direction. When there is a negative relationship between the two variables, one increases while the other decreases. Correlations are measured from -1 to +1, where -1 indicates a strong negative relationship, and +1 indicates a strong positive relationship. In statistics, there are several ways to measure correlation and Pearson correlation is one of them. The Pearson correlation is denoted by the letter "r" that measures the linear relationship between the two variables. When r is above 0 and has a "+" sign, then the relationship is positive. When r is below 0, the sign will be "-" indicating a negative relationship. If r = 0, then there is no relationship between the two variables. If there is no line between the data points on the graph, then there is no linear relationship. In order to get an accurate relationship and the accurate strength of r, any outliers should be removed Magiya (2019). Any outliers in the data can potentially make a difference in the strength of the relationship between two variables. Another factor that should be taken into consideration is that, the data should be normally distributed. A normal distribution of both variables can be seen when the values show a bell curve Magiya (2019). To ensure that there is not a mistake in the amount of values that were collected, each value that you put in from one variable should have a paired value from the other variable. This will accurately show if there is a relationship between one variable and the other. To calculate r, a computer program called SPSS can be used as well as a TI 83 calculator. To find the r in SPSS, begin by clicking Analyze, then, Correlate and then Bivariate. According to Kent State University Libraries (2021), "In the Correlation Coefficients area, select Pearson. In the Test of Significance area, select your desired significance test, two-tailed or one-tailed...flag significant correlations. Click OK to run the bivariate Pearson Correlation." By clicking ok, you are able to retrieve the value of r in the outputs.
What does "Sampling" in research mean? What is a sample versus a population? What is a representative sample and why is it important to have one?"
Let's iron some of this out. When you say "150 participants" this insinuates 150 people in the sample. I think what you mean is there's 150 in the population. If 50 are chosen, then those 50 comprise the sample. Your example about college students is a good one, but it's important to note that we can't make an assumption about the representativeness of a sample without knowing the population of interest. In this example, if your population of interest was American college students, then indeed, there would be great representation in this sample.
What is a Two-way ANOVA? How many IVs are involved? What type of design requires at least a two-way ANOVA (could be three, four, etc.)? That design tells us about main effects and what?
Let's make sure we identify some important vocabulary words here. The type of design that requires atleast a 2-way ANOVA is a factorial design. A factorial design tells us about main effects and interactions. An interaction is like what you described in your last sentence about high plant density and fertilizer type 2. A main effect refers to the effect of an IV on its own (i.e. the effect of plant fertilizer or the effect of plant density). An interaction refers to one level of an IV having different effects at each level of a different IV. For example, there may be no difference (when it comes to crop yield à the DV) in fertilizer types 1 & 2 when plant density is low, but a BIG difference (when it comes to crop yield) in fertilizer types 1 & 2 when plant density is high. This is an interaction because the effect of plant density differs depending on what type/level of fertilizer you are talking about.
What is eta squared? What does it tell us? What is its symbol?
Note that eta-squared is not the proportion of variance (total), it's proportion of variance accounted for by the effect/IV. Eta squared, with the symbol η2 , basically shows the percentage of variance from the independent variable n^2=SS between / SS total How do you find eta squared? You use your data from your ANOVA table in SPSS. After performing the necessary steps to complete an ANOVA table, you can find eta squared by locating the corresponding data in the table. Let's iron out your example a little bit. Let's say you are examining your food intake to determine what's contributing to your heartburn. OK, so heartburn is your dependent variable (this is what you are measuring). Perhaps in particular you are looking at the effect of bacon (this is your independent variable). So, to find eta-squared for bacon, you would need to determine the proportion of bacon variance to total food variance. How much of the variance in heartburn "scores" is accounted for by bacon?? With our made up numbers, we get 1200/1400 = .857 or 85.7%. So, bacon accounts for 85.7% of heartburn "scores."
Describe the criteria for having a true experiment. What do experiments show that other designs cannot
Note that experiments are the only research designs that allow us to study cause/effect. True experiments are the best, though researchers can draw tentative causal conclusions with quasi-experiments. Also, random selection is not a criterion for a true experiment. Random selection is helpful in any study, but it's not necessary to conduct an experiment. Finally, the criterion of "control" refers to controlling for extraneous variables; it does not refer to a control group. Most well-designed experiments have a control group, but one is not required.
Describe the following in words: Range Standard Deviation Variance
Note that standard deviation is not just how much a single number/score differs from the mean, but it's the average distance from the mean for an entire set of numbers/scores. Variance, then, is the average squared distance from the mean for an entire set of numbers/scores. A few other things to provide some more depth -- The range is the indicator of variability most affected by outliers because its calculation only includes the highest and lowest scores. Comparatively, standard deviation and variance include all scores from a data set. Standard deviation is the indicator of variability that you are most like to see used in a journal article.
What is an Extraneous variable? (characteristics of?) Provide a few hypothetical examples.
Note that there are 2 different kinds of extraneous variables: confounding & nuisance. A confounding variable varies systematically with the independent variable, whereas a nuisance variable does not. A confounding variable threatens internal validity, as it can provide an alternative explanation for a study's results (something other than the independent variable). A nuisance variable can obscure the effect of the independent variable; that is, you might not find an effect when really there is one.
Describe the Steps of the Scientific Method. Why do we use this method?
Observe > Ask Question > Make Hypothesis > Test Hypothesis >Accept or Fail Hypothesis >Make Conclusion *Note that it's not necessarily the case that if a hypothesis is refuted, a different one should be posited. It could be that the hypothesis or the related theory should be revised. It could also be the case that how the research study was executed should be revised. Also, it is true that by using the scientific method, studies can easily be repeated -- and this is a great point. Equally as important is that we use the scientific method as a way of acquiring knowledge in order to avoid the types of biases that we would run into if we relied only, say, on our intuition.
What are the three types of t-tests and what do each test between?
Recall that a t-test investigates the difference between 2 means. (You will use an ANOVA if you have more than 2 means to compare) Indeed, there are 3 different types if t-tests: (1) independent samples (2) paired sample (3) one sample. The independent samples t-test, as the name suggests, is going to test the difference between 2 groups that are not related to one another. For example, perhaps you want to study the difference in anxiety scores for a group of avid gym-goers and a group of individuals who do not exercise regularly; the type of t-test you would use would be an independent samples t-test. On the contrary, you use a paired sample t-test when the 2 groups are paired/related in some way. For example, you could use a paired sample t-test to investigate the differences in scores for participants' pre-test and post-test. Though the scores were collected at different times, they were collected from the same people -- therefore, they are "paired." A one-sample t-test allows you to compare the mean of a sample to the mean of a population. For example, for my master's thesis, I collected my sample's mean score on the Narcissistic Personality Inventory. I then conducted a one-sample t-test to compare my sample mean with the population mean -- how did they fare against the wider population (this is what I wanted to know)?
Describe the following sampling techniques:
Regarding random selection, each individual in the population has an equal chance of being selected for the sample. Note that random selection does not always involve sub-groups/strata. For example, simple random sampling does not make use of sub-groups/strata. Also, with stratified random sampling, the same number of individuals comprises each stratum/sub-group, but this is not necessarily the case for cluster or multi-stage sampling. A final note about random selection: whereas convenience sampling threatens external validity, random selection is good for external validity! The sample is likely to be representative of the population if it is randomly selected. Important note: convenience sampling is the most common type of sampling used in the behavioral sciences.
Distinguish between "statistical significance" and "effect size". What do they mean? What is statistical significance affected by that effect size is not?
Statistical significance and effect size both deal with statistical validity in both correlational and experimental research. When the results are statistically significant, it means that they are unlikely due to chance and the observed or recorded difference is due to the independent variable (Bankhead, 2020). Statistical significance is determined by the p-value, which should be smaller than the alpha level, which is usually set at 5%. Therefore, a p-value less than 5% or .05 is deemed to be statistically significant, and there is a low probability that the results of the experiment are due to chance (Bankhead, 2020).
What are "faulty" methods of acquiring knowledge that we want to avoid? Why is each of them faulty? (ex: tenacity, authority, intuition, personal experience outside of scientific context). Provide an example of each to help strengthen your arguments.
The faulty methods of acquiring knowledge are intuition, authority, rationalism, and empiricism. On some occasions, the scientific method can also be seen as faulty. Using intuition to acquire knowledge refers to letting your emotions guide you to your answers. Cuttler (2017) mentions, "...intuition involves believing what feels true." This method of acquiring knowledge can lack scientific reasoning. An individual's intuition can be wrong and may lead an experiment in the wrong direction. For example, your intuition may cause you to love someone the first time you meet them. Your intuition may lead you to believe that an individual is a great person, but there is no significant evidence that an individual is a good person. There are also more common methods of acquiring knowledge that may be seen as faulty. Authority being used to acquire knowledge is common. Cuttler (2017) states, "This method involves accepting new ideas because some authority figure states that they are true." The issue with this method is that the authority figure may be wrong. There may be no evidence or reasoning behind the information that is given. For example, as a child, my parents led me to believe that having the light on in a dark car was illegal. Even though I am now aware that this is not true, this idea still runs through my head. The next method that will be mentioned is rationalism. Rationalism refers to arriving at a conclusion based on a specific premise (Cuttler, 2017). For example, if you were told as a child that a daisy is yellow, you would assume that a daisy is yellow without even seeing a daisy. An issue that may occur is the premise may be wrong. This will cause your conclusion to also be wrong. Unlike rationalism, empiricism includes personal experiences and observation (Cuttler, 2017). This may cause individuals to only believe what they have seen or experienced. The problem with this method of acquiring knowledge is your experience can be limited. You cannot see and experience everything that is in the world. For example, some may not believe that other planets exist because they have never seen another planet. The last method of acquiring knowledge is the scientific method. You may be familiar with this method, but you may not be aware of the problems this method may cause. Cuttler (2017) describes the scientific method by stating, "The scientific method is a process of systematically collecting and evaluating evidence to test ideas and answer questions." The issue with the scientific method is that it cannot answer all questions and it can be time-consuming (Cuttler, 2017).
What is a One-way ANOVA? How many IVs are involved? How many groups/levels can you test between/among? What is the symbol?
The one-way ANOVA is an analysis of variance and it is used to draw inferences of differences between population means. In conclusion, an one-way ANOVA is helpful to see if there are any differences between two or more population means. It has no limitation on the number of means (groups/levels). A one-way ANOVA only needs one independent variable. The null hypothesis indicates that all means are equal, while the alternative hypothesis indicates that at least one mean is different (Howell, 2017).
Distinguish between Type I and Type II errors in research.
The probability of committing a Type I error is known as alpha (that's right, the alpha you know and love). The probability of committing a Type II error is known as beta. So, when you set your alpha (typically .05), you are setting your Type I error rate. You might be thinking, "Well, heck, why not just set that error rate super low! I don't want error!" Well, the problem with being super strict with your alpha/Type I error is that you then increase your beta/Type II error (failing to rejecting the null hypothesis when you should). Let's go a little further. I mentioned above that Type II error is known as beta. Type I error is power (the chance of rejecting the null hypothesis when you should because its false). The higher your power, the better. Note that power ranges from 0 to 1. Say your power is .8. Awesome! This means that there is little chance of rejecting the null hypothesis in error -- there is little chance of us missing a real effect (of our independent variable). Conversely, say our power = .2. Boo! This means that there is a great chance of us rejecting the null hypothesis in error and therefore missing a true effect (of our independent variable).
What does "Validity of a Measure" mean, in general? What is content validity? What is face validity? Explain Convergent and Divergent Validity.
The validity of a measure simply asks if the created measure actually measures what it is supposed to measure (Marek, 2020). In short, does the measure you created actually do its job within it's guidelines. When asking if content validity and face validity is valid, question if it looks like a good measure When asking about convergent and divergent validity, ask if the pattern makes sense. Content validity is "the extent to which a measuring instrument contains representative sample of items about the knowledge, trait, or behavior being assessed" (Marek, 2020). To successfully validate content validity, it should contain everything that your theory says it should. Your items should also should be checked against other topics in the overall domain of the discussion to check if it truly represents what it should be. Face validity is looking at a measure and seeing if it appears to measure what it should. This is just a general judgment, meaning that there is no official systematic way to test face validity. That means that face validity can vary from experiemnt to experiment, depending on what the researcher's perceive. Convergent validity states that "a measure should correlate strongly with other measures of the same construction" (Marek, 2020). This says that two separate tests, each having the same objective as the other, should be similar to one another. For example, there are two depression scales used in an experiment. When the two scales are compared for similarities of their content, there is a strong positive correlation. This means that both scales contain similar content to themselves. Divergent validity, or discrimination validity, states that "a measure should correlate less strongly with measures of different constructs" (Marek, 2020). This means that two separate measure, each having different objectives from the other, should not be similar to one another. For example, there is a depression scale and a physical health scale used in an experiment. When the two scales are compared for similarities in their content there is a weak positive correlation. This means that the scales did not have a lot of common content with each other.
What are the Major Ethical Themes in Research with Humans? Explain them in plain words
There are seven main Ethical Themes with Human Subjects the first is Respect for Persons which means signing people up for research without bribes or coercion. The second theme is Beneficence and non-maleficence this Is basically to protect the participant. The benefit of the experiment must outweigh the risk or harm associated with the experiment. Justice is third, this means to be fair between all participants and not treat them differently. Informed Consent is the fourth theme and it is needed in order to conduct research, and is basically how you tell the participant what is happening in the experiment. Confidentiality is protecting your participant and their responses or results to the experiment. Integrity is a theme made to help improve internal validity as well as keep data accurate. Conflict of Interest is where a particular researcher personal endeavors might overlap with the experiment tainting or leading to miscalculated results.