# Combination of non probability and probability sampling

#### Iron21

##### New Member
Can I use purposive sampling to select customers from a store of population size 200. And then employ simple random sampling to randomly select 100 participants from the 200 customers as my sample? My study is quantitative and wanted to know if they will be a high degree of biasing and inaccuracy in the results?

#### hlsmith

##### Less is more. Stay pure. Stay poor.
I think you may need to provide more details and state who you want to generalize the results to.

Thanks.

#### Iron21

##### New Member
My study is based on finding the impact of customer satisfaction on firm revenue. I want to generalise the results to only the long term consumers. As they have been my long term customers and a considerable amount of information could be gathered from them. But I'm worried about just using simple random sampling only as that will include the short term customers as well in the sample.

#### joeb33050

##### Member
Purposive or any non-random sampling exercise forms a distribution of a subset of the original distribution. The distribution of women basketball players is a subset of the distribution of basketball players.
If customer satisfaction is related to long term; then do regression analysis on length/$spent or some index of term vs revenue. EX visits/month vs$ revenue/year. The analysis will lead you to further analysis, then to a conclusion. Don't worry about bias until the conclusion is reached.