in statistics there are some terms that frequently come up two of these are a sample and a population in this video tutorial I'll explain what a sample and a population is in terms of statistics and I'll give you an example of each so the easiest way to understand the concept behind what a sample and a population is is to simply use an example let's say we are performing an experiment where we are interested in people who have a disease and this disease is called disease X it's estimated that over 1 million people in the world have disease X the population in this case is every person in the world who has disease X as scientists it would be extremely hard to almost impossible to recruit every person in the world who have disease X for example some people may decline to be recruited in a study and some people may be in a different country altogether too the study itself so what's the next best thing well to get around this we perform an experiment using random selection of people with disease x-saber recruited 50 people in the study with disease X this group right here is known as a sample the sample usually contains a smaller number of people compared with the population so why do we do this well as I've mentioned it is difficult to recruit everyone with disease X so the best approach scientists use is to recruit a randomly selected sample that is then used to make a general conclusion about the population let's say we measured the life expectancy of people with disease X in our sample we calculated that people in our sample with an average 51 years this measure since it has come from the sample is known as a statistic and the mean symbol for a statistic is x-bar usually we do not know the actual value the average life expectancy in this example for the whole population but let's imagine we know what the average life expectancy is for the population of disease X and this is 56 years this measure since its come from the population is known as a perimeter and the mean symbol for a perimeter is mu notice how the value for life expectancy in our sample is different to that of the actual population why might this be well there are many reasons for this but two common explanations are sampling error and selection bias something error occurs when the sample by chance has different characteristics to those in the population and this is because the sample is not the whole population for example just by chance we happen to recruit those people with disease X who also exercise less often and ate a poor diet compared with those in the population since these factors also linked to the statistic the life expectancy they may result in the life expectancy being lower than that is expected and so these factors may contribute to sampling error selection bias occurs when the sample is not selected in a true random fashion instead certain people may be preferentially recruited for example let's say for our study we advertised it through Facebook this means only people who use Facebook will be recruited whereas those who do not have a Facebook account will be ignored this is known as selection bias so now you understand the difference between a sample and a population the population contains all the observations whereas the sample contains just a selection of observations from the population the sample is used to make general conclusions about the population the statistic measured in a sample is often different to the parameter in the population and this is due to multiple factors to include sampling error and selection bias did you like this video be sure to give it a like I'll leave a comment and don't forget to subscribe to be notified when a new video is added