Download Survey Design and Analysis Techniques and more Exams Research Methods in Psychology in PDF only on Docsity!
Survey and Correlational Design (RM)
open-ended questions - With ________-items people can give responses that are impossible to summarize. What you can do with __________-items is create categories that peoples responses can go into. For example, we can have different categories so if a question is totally off chart, we can put them in places where they belong. open-ended questions - few restrictions = maximal response (advantage) open-ended questions - Responses may be impossible to summarize At minimum categorization requires IRR Potentially difficult statistical analyses Affected by ability and willingness to respond (Disadvantages) open-ended questions - Another disadvantage is that there is no guidance on how long the question might be. The person might not want to type or write the question which can lead to questions being left blank. Also, people might not have the ability to respond to ________ item questions. To avoid this, the best thing to do is to create a close-minded design. open-ended questions - Cat 1- Extreme Love for Football Cat 2- Moderate Love for Football Cat 3- Miscellaneous Cat 4- Questions Totally Off Chart The one issue we run into is that the person making the survey cannot make the categories themselves because they can bring bias to the issue. However, we can find one other person to make categories after reading the design and see how much they correlate to your categories to reduce bias. Then we calculate interrater reliability, this would allow other researchers to see you tried to reduce bias.
closed-ended questions - Easy to summarize and analyze Can get quantitative data Blank category option These are advantages of closed-ended questions - May limit expression closed-ended questions - You can get ordinal or ratio data for stat analysis You could provide a blank category option where people can write in their own answer if they do not see their answer choice closed-ended questions - You might limit expression, so blank category option can help them with what they really think. rating scale - On the second scale, we have a scale with no pictures and this could be used for people who can read. You could put an X where you feel on this scale. Lets say for people who can read. You could put an X where you feel on this scale. Lets say you put an X between very and somewhat hungry. Than after that you put an O after dinner to show if you are still hungry. Researchers measure the distance between the X and O after looking at the scale. rating scale - you can use interval data. We can measure the distance between X and O in one of the scales. One on scale there is 2 point intervals which for example would 1-2 would be the same pain level. rating scale - When you use interval data you have more option for stat analysis ____________ are easy to understand
- How often do you visit a dentist?
- How much time do you spend on a visit to a dentist? double barreled questions -
- How satisfied are you with your pay and work environment? Should be
- How satisfied are you with your pay?
- How satisfied are you with work environment? double-barreled questions - ___________ questions are one of the most common surveying mistakes, but also probably the easiest to fix. You correct a _____________ question by separating the combined question into two distinct questions. This is how you fix this issue. put questions together to avoid confusion - If we are focused on academic performance and self-esteem, you want to clump these questions. Due to this you want them to be in the same context, clustering them together can help with avoiding confusion. sensitive - A _____________ question should be in the middle, because you do not want to start the survey that you do not know with sensitive questions because they might not want to answer them. demographic - You put ___________ your questions at the end, because these are just trying to figure out how old they are or where they come from. So if someone skipped a question you would want them to skip a question like this instead of a regular question involved.
general questions - if we are asking people to rate survey on a basis 1 through 10 on self-esteem we must be specific. How do you feel about self-esteem after winning a 5k, or getting good grades, or after a break up. mail delivery - Non-threatening, and no one sees you. You do not have to interact with people, these are the only primary advantages to sending through ___________. mail delivery - Would be expensive, printing costs, mail costs, and other things Paper surveys have low response rates, which is low as 15 percent. disadvantage mail delivery - We could use a good cover letter and tell them why it is important, also tell them that their data is important improving mail delivery - We could always provide a free gift with the survey, like have a coffee on us, or give them a small coffee gift card mail delivery - Give advanced notice, reminders, and thank you letters for the interest of for completing survey. Send follow ups after sending the survey out. improving online survey - You have sampling flexibility with the internet, you might have a much quicker idea of who will respond to survey. Immediate feedback to who will finish the survey. You can see that there
Best controlled groups, you can send these people their own personal link where only one person can take the survey even if the link is copied and pasted phone survey - You can have automatic digit dialing More people who are older will be easier to get hold off on this method Researchers can get through to people easier This is great for random sampling Can do it from home or office phone survey - You must have someone on the phone to do the survey Your bias could affect the results of the study. If you have an individual who is in support of what the researcher is looking you could get more influential responses from the interviewer or vice versa where the interviewer can influence response from participant disadvantage phone survey - Researchers may rephrase questions to get better responses If people think you are trying to scam them, it can really tax you emotionally disadvantage interview survey - Gold standard Highest response rate You can develop report, people get to know you better when it comes to this situation You get more sincere or accurate dispositions from people You can go in more depth with these types of surveys advantage interview survey -
You are much more likely to get more person questions and follow up with secondary questions. For example, have you been assaulted before, than you can follow up with how long ago was it. You can pursue topic in much more detail when it comes to this interview survey - An in-person interview will take much longer than any other interview method People can pick up on bias. These are all indicators that you have a particular point of view. Need good training so we can minimize bias in the data when doing personal interview disadvantage correlational research - Noticing associations between variables. The goal is to determine the strength between the relationships. They do not imply cause and effect. We cannot manipulate the elements in the part of the design. There are two many variables involved which means no cause and effect. correlational research - We can run a survey to see your interest in psychology and run a survey looking at your GPA in this major. We can look to see of higher interest in psychology is correlated with high GPA. correlational research - We might be interested to see if genetics is related to IQ. We can collect IQ data from children and collect IQ data from their parents, we would see how correlated the parents and children IQ were. correlational research - Could be interested in students attitude while in university, I might think that this impacted by what university they go to. We could do a survey to have students rate their university experience and also record the size of the school they are at. We could give this survey to many universities, we can see if university size and attitude are correlated monotonic spearmen correlation - a curvature relationship or undirectional relationship
This statistic is incredibly valuable, it tells us how much variance is accounted for by the correlation. correlational research - No manipulation over variables, drawback to __________? correlational research - Does not imply cause and effect. 0 - r= 0 the r^2 variance would account for _________? .16 - r=.4 the r^2 variance would account for ________? .64 - r=.8 the r^2 variance would account for _________? .1 - r=.2 the r^2 variance would account for _________? correlation design - If we know the weight of the bowling ball impacts how fast it will roll down the hill, we could have a good idea of how fast the bowling ball will roll even if we do not know the speed beforehand. We can make an accurate prediction. inner-rater reliability - You can calculate ___________ reliability be seeing how correlated to raters reponses are in a study. The higher correlation, the higher agreement between raters. This can help reduce bias and help with reliability in correlational research. theories in correlational -
We can theorize self-esteem would allow positive outcomes in the classroom. What we could do is do self-esteem measures and see if there is a high positive correlation between positive outcomes in the classroom. correlational research - There are challenges with ______________, when we have a strong correlation often times people take it too far and try to establish cause and effect. Like self-esteem causes positive outcomes in school. third variable problem - You could see that people with birth control might own toasters Senators in Texas might pick up on this and decide to hand out toasters instead of having sex ed program So in this situation we can see that SES could impact the individuals on birth control to own a toaster You need money to buy birth control and toaster use, these are not related, but it was SES driving both of these outcomes. third variable problem - Private school shows higher correlation with scholastic ability Public school shows correlation with lower scholastic ability It does not mean public school means lower scholastic ability SES another third variable can influences where the child goes to school and can have an impact on on things like this. However, if one parent makes 80 grand with a kid going to private and the other parent makes the same with 80 grand a year, there should not be a difference in ability If you are higher SES and scholastic ability, the kids more likely have more opportunities, parents might have higher education level and other resources to help their children Kids from lower SES might not have all of these opportunities that the other kids have. If the mom is working three jobs to keep ends meat on the table she will most likely have a hard time helping the kid. third variable problem - Could the third mechanism be causing or impacting the two other responses. survey design -
A survey researcher chose to administer a survey using the Internet. This sample likely overrepresents people who own a computer and underrepresents people who don't own a computer. This problem in the researcher's survey is best described as loaded question - An emotional laden term is called an double-barreled questions - A survey question that asks, "Are you worried about the economy and getting a job?" is an example of a: open ended question - Please list your favorite cereals is an example of an Pearson correlation - used with two quantitative variables only appropriate if data is related in a linear fashion regression analysis - Use of one or more quantitative independent variables to explain or predict the values of a single quantitative dependent variable. multiple regression - It is used when we want to predict the value of a variable based on the value of two or more other variables multiple regression - involves two or more independent or predictor variables negative correlation - If the pattern of dots on a scatterplot runs from upper left to lower right what does this indicate? positive correlation -
For her senior research project Jerri collects survey data on the number of hours spent on social networking and grade point average (GPA). She creates a scatterplot to graph her results. If the pattern of dots on her scatterplot goes from lower left to upper right, what would this indicate? strong negative correlation - strong negative correlation correlation coefficient - +1 and -1 is a perfect? pearson correlation - When calculating a correlation coefficient, ________________ can only be used when the data are related in a linear fashion. multiple regression - How does _________________ differ from simple regression? It includes multiple independent/predictor variables multiple regression - For example, you could use _____________ to understand whether exam performance can be predicted based on revision time, test anxiety, lecture attendance and gende multiple regression - Alternately, you could use ______________ to understand whether daily cigarette consumption can be predicted based on smoking duration, age when started smoking, smoker type, income and gender.
