Class Notes (835,581)
Canada (509,259)
Psychology (7,782)
PSYB01H3 (260)

Chapter 5- Sampling and Servey Resarch.doc

8 Pages
Unlock Document

David Nussbaum

Chapter 5- Sampling and Survey Resarch Substantive theme : Happiness • basically do a survey bout happiness Selecting research Participants • dff between survey and expiral ( in the lab) • Survey asks spec question to someone’s feelings and u can do an manipulation if u Δ the question but that’s it … like if u ask 2 diff questions to diff groups and then u compare the results but thus manipulation is rare.. Ur usually trying to find the consensus of certain question and u want as many ppl as possible opposed to choosing certain amount of ppl and doing a study on them ( TAPS) , u can survey ppl for free on survey monkey ( if u want extra add-ons u pay them but just for general its k ) • 2 basic question that one has to as if ur doing surveying ppl Sample Planning • To plan a sample, you need to answer two questions: • 1- From what population will you select the cases?→Like what part of the pop will u ask .. U might wanna sample ppl outside Ontario f u already know what Ontarians think • 2- What method will you use to select cases from this population? o How do u go through surveys o U may not want everyone to survey ( 3-5 years)- ur obviously gonna do a survey in eng so maybe u dunt want the foreign ppl , what about ill ppl in hospitals so ur not gonna try to get every last Canadian so u need some method to select what cases ur gonna target o Then u got to say clearly what pop u want to study/ survey Define Sample components • once we know what pop were gonna study we need to develop procedures for drawing a sample from that pop that rep of the pop • State clearly the population that is actually under study • population – entire set of indiv or other entities to which study findings are to be generalized • Elements / elementary units- Individual members of the population whose characteristics are to be measured • Sampling frame- List of elements in the population • rep sample – sample that looks like the pop from which it was selected in all respects potentially relevant to the study • Eg if ur doing market studies ur not gonna wanna use teens if ur gonna study like who buys a car or not • sample generalizabiltiy depends on the amount of sample error- diff between the characteristics of a sample and the characteristics of the pop from which is selected o larger the sampling error , the less rep the sample so less generalizabiltiy o U can’t get everyone in the population so sometimes the sample is not representative of the whole population but its rep of a small portion of the whole so this is really representative of the whole – so what u want to do is ↓ this error Estimating Sampling error • Inferential statistics- Tool for calculating sampling error o Our sample is one of many that could be drawn from a population o Sampling distribution - Represents all of the possible samples that we could have drawn, Many are “normal” o “Normal” shape produced by random sampling error o Most frequent value is identical to corresponding population parameter • random sampling error ( or chance sampling error ) – diff between the pop and sample only due to chance factors ( random error ) not to systematic sampling error • sample statistic – value of a statistic, like the mean , computed from sample data • population parameter- value of a stat , like mean , computed using the data for the entire pop o sample stat is an estimate of the pop parameter Sampling Methods • Probability sampling methods: let us to know in advance the likelihood of selecting element of a pop for the same , Use random selection, No systematic bias o Virtually all studies employ random sampling • Non-probability sampling methods: Do not let us know in advance the likelihood of selecting each element and thers bias • probability of selection – likihood that an element will be selected from the pop for inclusion in the sample • natural to confuse the concept of random sampling Probability Sampling Methods • 1- Simple random sampling- Procedure that generates #s or identifies cases on the basis of chance o Every member of the population under study has an equal chance of inclusion in the study o random digit dialing – when organization conduct phone surveys often draw random samples using automatic procedures ( machine dials random #s ) • 2- Systematic random sampling- First element is selected randomly from a list and then every nth element is selected o Must watch out for periodicity- the sequence varies in some regular periodic pattern • 3- Stratified random sampling- All elements distinguished according to a characteristic (forming strata) and Elements are sampled randomly from within the strata o Can be Proportionate or Disproportionate o proportionate stratified sample – ensures that the sample is selected so that the distribution of characteristics in the sample matches the distribution of the corresponding charec in the pop o disproportionate stratified sampling - • 4- Cluster sampling- Useful when a sampling frame is not available o Random sample of clusters is drawn o Elements within each cluster are randomly sampled o Ike u choose certain geographic area and u do random sampling within that cluster Nonprobability Sampling Methods • Also called convenience, or availability sampling • Quota sampling = intended to overcome most obvious flaw of availability sampling that the sample will just have whoever and whatever is available o quotas are set to ensure that the sample represents certain characteristics in proportion to their prevalence in the population • Welcome to psych... Psyc is the study of 1 year undergrad students but if u looks at majority of studies done in units and labs the participants are undergrad studies (TAPS) …. Other ppl who aren’t really in unit and psyc courses dunt really do these tests … usually first year undergrad psyc courses • Who says that it implies to ppl who have never taken first year psyc courses… cuz the major ppl being tested are the undergrads … does it generalize??.... We dunno if its tested a lot and proff is given then maybe it is • Point s that this is convenience sampling … like if we go in the city and just ask random ppl it wud take a lot of time and its inconvenient … we just take ppl that we know that are like alive and around us ( taps u sign up for a time and time slot and u do it ) Writing Survey Questions • 1- Avoid confusing phrasing- Shorter words and questions, Avoid double negatives o Avoid double-barreled questions (asking two questions in one) o Add filter questions- basically something like that will exclude the person from tabulating results within the overall data o Eg if u ask about break time in this class but the person answers there not in the class so then u just remove that question for that person cuz it’s not necessary o It doesn’t exclude them from finishing the survey o create skip patterns so it’s like if answers yes to question 14 do the following – that’s the contingent question o Make things as clear and simple for ppl to respond accurately o Malingering is diff... It exaggerating for monetary reason ( they have personal gain) • 2- Minimize risk of bias • 3- Avoid making disagreement or agreement disagreeable • 4- Minimize fence-sitting and floating- Do not force neutral individuals( fence sitters) to choose between opposites o Let’s say u can have a question about capital punishment ( do u believe it or not ) – there are ppl who aren’t comfortable with it butt thers no possibility for them to Δ or can kill
More Less

Related notes for PSYB01H3

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.