- CFA Exams
- CFA Level I Exam
- Topic 1. Quantitative Methods
- Learning Module 6. Hypothesis Testing
- Subject 2. Null Hypothesis and Alternative Hypothesis

###
**CFA Practice Question**

A social psychologist reports that "in our sample, ethnocentrism was significantly higher (p < 0.05) among church attendees than among nonattendees." This means ______

B. the observed differences between church attendees and nonattendees account for all but 5% of those sampled. These results are quite meaningful and should be investigated further.

C. if there is actually no difference in ethnocentrism between church attendees and nonattendees, the chance that we would observe a difference as large as or larger than we did is less than 5%.

A. ethnocentrism was at least 5% higher among church attendees than among nonattendees.

B. the observed differences between church attendees and nonattendees account for all but 5% of those sampled. These results are quite meaningful and should be investigated further.

C. if there is actually no difference in ethnocentrism between church attendees and nonattendees, the chance that we would observe a difference as large as or larger than we did is less than 5%.

Correct Answer: C

Recall that the P-value is the probability, computed assuming the null hypothesis is true, that the test statistic would have a value as extreme or more extreme than actually observed.

###
**User Contributed Comments**
14

User |
Comment |
---|---|

surob |
Can someone help me understand it? thanks |

randa |
As I understand it from the thoery. This is a statement of a one tailed test. A one tailed test has to be stated as H0: x>= or x<= and H1: X STRICTLY > OR STRICLY <. As such, you have to such for the answer that allows to give you an H0 being greater or equal or smaller or equal. Answer d sis the only answer that provide such a format. |

studyprep |
All it is about accepting or or not accepting the Null hypothesis. 'C' is saying we can safely not accept the null hypothesis, because p<0.05, i.e. there is not much significant difference in ethnocentrism among both type of groups... ---please correct me if someone gets plausible explanation. --- --- I think, here we do not need to know whether it is a two tails or one tail test. Tails are established particularly when we look for numbers, such as portfolios returns or number of patients. |

johnowens |
H0: ethnocentrism was significantly higher among church attendees than among nonattendees. p < 0.05. So the probability of ethnocentrism being higher for attendees, is only 5%. |

johntan1979 |
That's wrong. Significantly higher, reject null. Null: No difference in ethno |

sgossett86 |
these comments didn't help.. |

PJMOHAN |
yup!! didnt help at all |

KrzysztofW |
From this sample we Do know that ethnocentrism was SIGNIFICANTLY higher among church attendees than non attendees. With p<0.05 it's just mean that probability of Ho being false is just less than 5% So there is a less than 5% chance that , even though according to the sample ethnocentrism was SIGNIFICANTLY higher among church attendees than non attendees, in population there is no difference at all |

floydtrend |
Ho: Church goers not more ethnocentric than non-goers H1: Church goers more ethnocentric than non-goers P-val. is 0.05 or Level of significance is 5% =>That Ho is true, but there is a 5% chance of H1 being possible |

Olesya_CFA |
Johntan1979 gave the best answer in my opinion. Thank you. |

dbedford |
Null Ho: People of the same race that go to church tend to spend more time with each other more than people that dont go to church who dont tend to racially congregate. P<.05 means that the chance of them being wrong based on their calculations was less than 5% C. says if there is no difference between church members of the same race hanging out with each other and non church attendees, we didnt see it and the chance that we are wrong is less than 5% |

dbedford |
sorry my bad i wrote race but ethnocentricity means people of the same cultural values, which makes more sense. |

shalie |
I think the key to this is that the p< .05 . |

mcbreatz |
Some of this may be technically wrong but I think of it this way and it helps me. Null or H0 was that there was no ethnocentrism and tested at a 95% confidence interval (1-.95= P.05). The fact that they found ethnocentrism is highly unlikely if H0 is true. So reject H0. Answer C is only one that has anything to do with rejecting a null H. |