Help
Options
Question
what is risk
click to flip
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't know
Question
what is an example of risk
Remaining cards (69)
Know
retry
shuffle
restart
0:02
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
COMM 365
In-Class Slides
| Question | Answer |
|---|---|
| what is risk | we know the possible outcomes and their likelihood |
| what is an example of risk | if we bet on a fair coin, there is a 50% chance of heads |
| what is uncertainty | we are missing information |
| what is an example of uncertainty | we dont know if that coin that has a 50% chance of heads is a loonie or a toonie |
| what is both risk and uncertainty | we are missing information about the outcomes and or their likelihood |
| what is an example of both risk and uncertainty | if we bet on a loaded coin there is an unknown chance of heads |
| what is no risk or uncertainty | we know the only possible outcome |
| what is an example of no risk or uncertainty | if we bet on a coin with two heads, there is a 100% chance of heads |
| we don't know whether customers use soap at night or in the morning a. risk b. uncertainty c. risk and uncertainty d. no risk no uncertainty | uncertainty |
| we dont know the chance a new soap will succeed a. risk b. uncertainty c. risk and uncertainty d. no risk no uncertainty | risk and uncertainty |
| we know that if we do not launch the new soap it will not succeed a. risk b. uncertainty c. risk and uncertainty d. no risk no uncertainty | no risk no uncertainty |
| we know that there is a 10% chance that the new soap will succeed a. risk b. uncertainty c. risk and uncertainty d. no risk no uncertainty | risk |
| exploratory research | defines problems and clarifies alternatives focus groups, depth interview, open ended surveys |
| descriptive research | describes market characteristics observational studies, surveys, secondary data |
| causal research | determine cause and effect relationship experiment |
| what is the usage competition benefit framework | usage occasion, competitors considered, and perceived benefit |
| types of survey methods | in person, by phone, by mail and internet |
| Sampling error | difference between population and sample due to chance variation reduce this error by increasing sample size |
| Population specification error | occurs when researchers don't know who to survey (1. sample pop is too narrow, wide, or different 2. incomplete or inaccurate sampling frame 3. sample procedure is inappropriate) |
| Facial recognition | can create targeted ads |
| Surrogate information error | due to difference between data needed and data sought ex. asking what big chain restaurant people prefer when you are trying to gather data on small ones |
| Non response error | due to difference between respondents and non-respondents ex. only people who have previously ate on uni blvd responding to the survey |
| interview bias | due to intentional/unintentional influence of the interviewer ex. they're wearing a tshirt advertising a product mentioned |
| response bias | due to intentional/unintentional misrepresentation of info ex. participants are embarrassed to answer certain questions truthfully |
| processing error | due to error in data recording, transcript etc. ex. respondents put A but surveyor writes D |
| types of measurement scales | ratio, interval, ordinal and nominal |
| ratio | ratio like 2/4 = 4/8 |
| interval | range like 2-1 = 7-6 like fahrenheit and Celsius |
| ordinal | numbers are assigned based on their order in a ranking formation like 0<1<2... ranked preferences, education levels etc. |
| nominal | percentages, frequencies like 0, 1, 2,... brands in a category, store types, etc. |
| natural vs contrived | observing behaviour as it takes place in its typical environment vs observing of behaviour within an artificial environment where other variables are controlled |
| disguised vs undisguised | the respondents are not aware that they are being observed, vs the respondents are aware that they are under observation. |
| Human vs machine observation | uses human observation to collect data vs uses machine observations |
| correlation | relationship between two variables |
| causation | the change in one variable produces an effect in another variable |
| three conditions for causation | correlation (evidence of association), time order (X occurs before Y), no extraneous factor (control for other causal factors) |
| X-O-R syntax | R individuals have been assigned at random to separate treatment groups X exposure of the test group to an experimental treatment O the processes of observation or measurement of the dependent variable on the test units |
| test units | are the entities to whom the treatments are presented and whose response to the treatments is measured |
| treatments | are the manipulated alternatives or independent variables whose effects are then measured |
| dependent variables | measures taken on the test units |
| Factorial designed experiments | used to study the effects of two or more variables at the same time |
| Factorial design table | *know how to use |
| primary data | collected for the purpose of the current project |
| secondary data | already collected and often published for some other purpose than the current project |
| backward market research approach | use a dummy table (all elements of an actual table without actual data) to conceptualize your analysis and understand what data is needed to be collected |
| Probability sampling | each element of the population has a know chance of being selected for the sample (random sampling) |
| Non-probability sampling | each element in the population has an unknown chance (convenience sampling) |
| simple random sampling | every member has an equal chance of being selected ex. all UBC students have a 1/5 chance |
| Sampling error | (95% sampling error = 2 srt(0.25 / n) 0.25 is the sample std n sample size |
| cluster random sampling | population is portioned into mutually exclusive clusters then do random sample to get clusters ex. break UBC students into ethnic groups randomly survey some students in each group |
| stratified random sampling | divide the population into mutually exclusive groups, then do random sampling ex. break UBC students into faculty randomly select a few faculties and survey all students there |
| p-value interpretation | If the null hypothesis is true the probability of observing the sample (test statistics) or something more extreme than (insert p-value). |
| conclusions | if p-value is lower than the sig value reject the null if p-value is higher than the sig value do not reject the null |
| null hypothesis | the ordinary state of affairs |
| alternative hypothesis | the opposite of the null and represents what we are testing for |
| pearson correlation coefficient | r<0 is negative linear correlation r>0 is positive linear correlation r=0 no correlation r= ± 1 perfect positive/negative correlation |r|= the closer to 1 the stronger the linear association |
| multiple linear regression | to understand how multiple independent variables are relared to dependent variables |
| simple linear regression | to understand how one independent variable is related tt o a dependent variable |
| correlation based recommender system | calculate similarity of ratings for Ed to other customers sort customers based on similarity to Ed choose a subset of x closest neighbour calculate predicted ratings using wgt avg recommend items with high predicted scores that haven't been rated yet |
| linear correlation (3 caveats) | 1. linear correlation only measures linear relationships 2. they measure how closely data is scattered across a linear line 3. when linear correlation is 0 there is no relationship between the variables |
| market research | Process to obtain info/data and make sense of it / make decisions to min risk and make profit |
| Consumer shopping behvaiour after covid19 | Less in person and more selfcheck outs |
| example of exploratory research question | what do you think of happiness or how can we improve our current product |
| example of descriptive research question | what gender are you or how much do you consumer apples |
| example of causal research question | does political advertising increase (cause) vote shares? |
| google golden triangle | In eye-tracking studies, researchers have observed that users tend to focus more on the top-left portion of the search results page, forming a triangular pattern that resembles a "golden triangle. |
| Application of facial recognition and targeted ads, pros and cons | Pros Insight into behaviours rather than reported behaviours No recall error risk Cons Ethical concerns Algorithm bias |
| Experimental designs | involves the specification of treatments (X) to be manipulated, test units to be used, dependent variables (Y) to be measured, and procedures for dealing with extraneous variables |
| Estimated Model: | An estimated model is a statistical model that has been developed using sample data from a population. |
| Population Model: | A population model, on the other hand, represents the true underlying model that would describe the entire population if it were possible to collect data from every individual or element in that population. |
Created by:
brookcoletta