hello students in this lecture we're going to be talking about sampling techniques so what is a sample well it's defined as a subset of a population sample is the select sample for the research the term population refers to a clearly defined group of cases from which the researcher can draw so you know your population could be athletes and then you pick certain athletes as your sample from that population so here are some key terms right so population like I just said the members of a specific group my example was athletes your target population the ones you ideally want to generalize accessible population people you can actually get to or data you can actually get to samples the subset of the population the sampling frame is the list of that population where you're able to draw a sample subject is a specific individual so sometimes we call him a case so that would just be one person you interviewed for example and sampling technique is the method you use to sample the population there's some steps here Define your target population who are you looking at select your sampling frame your sampling method your sample size collect your data and assess your response rate so your response rate means that if you try to talk to 100 people and nine out of 10 people talk to you then your response rate is 90% which is really good um it's not usually that high and online it can be lower than 5% so here are some sampling methods simple random sample well this method is every case or individual and the population has an equal probability of chance of being selected so it's a suitable CH choice if you have a small homogeneous people are the same easy to access stratified population divided into two or more groups called strata random samples are taken in proportion of the population from each subgroup so if you think about subgroups maybe it's an OCC Pati or a nationality the number of employees so the stratified means in layers you know like a cake cake has layers well you can have strata so in order to select we need to make the layers so the steps or stages involved in undertaking stratified sampling are one choose a variable gender occupation divide sampling frame into a discrete strata number each of the cases within each stratum with a unique number 1 2 3 4 5 6 7 8 9 10 select sample using simple random or systematic sampling we're going to talk about that in a second so systematic sampling researcher selects a sample from randomly generated list represents the sampling frame researcher selects every nth case so this is up to you you can say every second person or every third person or every 10th person the interval is determined by the size of the sample needed from the list until the desired sample size is reached so if you were sampling an nth case of maybe every third person and you wanted to get a sample of 300 people you know you'd be passing by somewhere in the neighborhood of 900 people um before you actually got your sample size that you needed cluster so think about that what do you do when you cluster you group things together so following this simple random sample of clusters is selected from the population so you have a big group which which we call our population and then you're going to have little groups which are your clusters type sampling is particularly useful when subjects are over a large geographical area saves time and money you know if you're going over an entire County or an entire state or even an entire country you might want to group people together into these clusters multi-stage sampling carried out in a number of stages so it gets reduced each stage so the process involves from moving from a very broad sample to a very narrow sample non probability is not everybody has the same uh chance of being selected for your sample quota researcher looks for specific characteristic like nationality and then you have predetermined characteristics so that your total sample have the same distribution as the wider population so if you know what the national the percentage of national background is in your state your region your county you know then you can actually do a quota that matches it 20% of people came from this nationality 30% of came from this and 50% came from this and you can do that snowball it's just like you think so if you put a snowball on the top of a snowy hill and you roll it and it starts rolling down the hill it picks up snow so what you're doing is that so you talk to a small number of subjects people you may know in maybe a business field that you're researching then after you've interviewed them you ask them well who do you recommend that I should talk to next right and so that's the snowball it's moving from that person to another person and you're collecting more and more based on these recommendations judgment is a method in which particular settings persons or events are selected deliberately that is your judgment in order to provide important information can't be obtained from other choices researcher includes subjects in the sample because they are deemed to Warrant inclusion so it's completely up to you right so it's your own judgment and it's viable just have to say so in your um write up convenience sampling convenience means it's easy right so marketing researchers use this a lot of times it's simple it's straightforward so if you want to maybe research shopping habits of young people decide to survey employees of you know a professional body some sort of shopping Association or something like that or you know just go and into a mall and talk to young people that you pass by and give them a survey about their shopping habits every person that you walk by because that's convenient it's easy how do you figure out what your sample size is going to be do you need five people 10 people or a thousand people well you know your research your due diligence look what if what if people in earlier studies done confidence that you need to have a new data level of certainty that characteristics that you collect in your sample actually represent the characteristics of the total population and a margin of error what can you tolerate the Precision you require for any estimate made from your sample you know and that delves into some statistics that we will get to later in the term choice of sample size comes down to the following type of analysis you're going to make how you going to research the data number of categories you wish to subdivide statistical techniques size of the total population and you can use formulas publish tables different things to help you there's actually tables out there that say if you're trying to do XYZ you need 200 people you can find these so collect your data data collection be categorized on the basis of primary and second secondary data methods primary is you collect it secondary is somebody else collected it wrote it up and now you can read what they have and use it assess your response rate and suitability again the response rate you know I gave an example of like if you talk to 10 people and nine people respond that's 90% response rate right it's defined as the number of subjects agreeing to take part and you write that up as a percentage all right that's it for this lecture I will see you next