Speaker 1: So, this is a video on creating surveys and how to make sure that you're asking the questions in the right way so that you avoid something that is called bias. So bias is an error that happens when you're collecting data. Sometimes it's intentional and sometimes it's unintentional. And it happens when you're trying to influence people to answer a certain way. So for example, we've all seen commercials where they'll say three in four dentists recommend Crest, blah, blah, blah. And they can't lie, that has to be a correct statistic. But you know, we don't know whether they only asked four dentists. We don't know whether they only asked dentists that were being paid by Crest to say that they were the best. So there's often a lot of bias or error that can happen when we're collecting data. And to make sure that we create the best survey possible, we want to try to ask questions in a way that does not lead people to certain answers. So here are some potential things that can happen with surveys. So for example, here, a bias survey. So a bias survey can result in data that does not accurately reflect the feelings of the sample. And we're getting answers that maybe aren't quite what people really feel like. And often, this happens because of how the question is written. And when we're creating our own survey, we have to make sure that we ask questions in a way that we're going to try to avoid bias. So here's an example question, how dumb is it to have a four day weekend in the middle of the term? So right away, we see that this has a negative connotation. Right away, they're leading the person into saying, oh, yeah, it is dumb, right? It's a leading question. Here's a different one. What are your thoughts on having a four day weekend in November? You're neutral. You're not saying whether you like it or not. You're just getting other people's thoughts. So again, here's a question, should concerned parents get flu shots for their children? Again, the language in this makes us think, oh, no, if I'm a concerned parent, I should do this. It leads people to a certain answer. So here's a more neutral question. Do you think children should be required to get flu shots? You're not showing your opinion. It's just a simple, what do you think? Another thing with questions that's interesting to consider is there are open ended questions where you can kind of answer anything. For example, what is your favorite color? There's maybe not a million colors, but there's a lot of colors that people could choose. That's called an open question. A closed question is more like this. Do you think children should be required, yes or no? Or you could even say, out of these five choices, what is your favorite color? So make sure when you're creating surveys, you have a combination of open and closed questions. Because sometimes if you have too many open questions, you're not actually able to make very good conclusions based on the data. Sometimes having more closed answers, closed questions will help you make better conclusions. That being said, if you have too many closed questions, then you're potentially having too many questions where you're not giving people the answer that maybe they want to be able to say. Here's another example. When do you like to hang out at the mall? So again, a little bit of a loaded question here, a little bit of a different way of saying it. What do you like to do with your friends? So again, if we look back here, we're sort of saying, you know, when you have free time, you're hanging out at the mall. Whereas this is more saying, what do you like to do? We're not mentioning the mall, we're not mentioning sports, we're not mentioning, you know, video games. We're just saying, what things do you like to do? Whereas this one is loaded. You're saying, you know, you're going to hang out at the mall. Here is another. How happy or unhappy are you with the lunchtime schedule and the quality of the food? So again, it's really not so great. So here, how happy or unhappy are you with the schedule? And then what do you think of the food quality? Compare that to this one here. So here we have like too many things happening at once. Here we've broken it down into two different questions because they're kind of two different things, right? I could be happy with the schedule, but not like the quality of the food. If I ask them all at once, so if I come back here and I have everything as one question, it's hard to really become clear on what people's opinion really is. Here's another one. Do you always shower before bed? Again, you're not really able to. So here is something. How many nights a week do you shower? And there's choices. This one's kind of like, do you always shower, yes or no, right? This one gives you some more choice. What is your favorite television show? And again, there's four options here. Again, it doesn't include everything. And I've already mentioned this. Sometimes you have to ask questions like this. If you leave everything too open too many times, you're not going to be able to make conclusions. But maybe another thing would be to say, what is your favorite TV show list for, and then have an option that says other. So there is a suggestion to improve the question. So you're still able to come up with conclusions, but you're able to allow people to have another choice. How would you describe the quality of the pizza at HSC? Excellent, great, or good? Again, not a great question because what if you don't like pizza? You haven't given people that option. So here is a better choice because we've added okay and we've added poor. So a lot to think about there in terms of questions. We want to try to ask neutral questions. We want to try to ask a combination of open and closed questions. Try to give people choice. Maybe if you don't want to give too much choice, at least have other as an option and try to avoid bias as best as possible when you are creating a survey.