Describe the process of NHST, tell the (five) possible outcomes and tell the likely reasons for each.
(Be sure to tell what this acronym means.)
1. First, you start off by either keeping or rejecting your null hypothesis. After that, the five possible outcomes are correct
retention, correct rejection, Type I error, Type II error, or a Type III error. Correct rentention and correct rejection are
alike in that both involve getting the correct answer and being supported by others. Type I error occurs when you
reject the null when you should not have, often occurs when your sample is too small. Type II error occurs when you
retain the null when you should have rejected it occurs due to having too small of a sample. Type III errors occur
when you do not correctly identify your variables and population. NHST stands for Null Hypothesis Statistical Testing.
2. When you reject the null you are saying that the p-value is less than .05. when you retain the null you are saying that
the p-value is greater than .05. There are 5 possible outcomes, the first is that you correctly rejected the null and the
second is that you correctly retained the null. These two reasons are due to the fact that you had good data, good
sample, good measures. The third is type one error. The fourth is type three error. These two reasons are due to that
fact that you may have had a false alarm (type I error) or a misspecification (type III error)because you didnt have
good measure, samples or data. The fifth reason could be a Type II error which could be because you didnt have
good measures, samples, or data but also because you didnt have a true experiment.
3. NHST stands for Null Hypothesis Significance Testing and follows a certain process. First a hypothesis is made and
the null hypothesis is formed that there is no difference or no effect in what we are studying. A sample is found that is
hopefully representative of the population we are studying. With the data we determine which statistic to use (r, X^2,
BG F, or WG F). If our p-value (or significant value) is less than .05, which means there is less than a five percent
chance of committing a Type I error, we reject the null hypothesis. If p > .05, we retain the HO:. The five outcomes
are: a correct rejection (reject the null and there is an effect in the pop), a correct retention (we retain the null and
there is no effect in the pop), we commit a Type I error (false alarm, we should an effect when we should not have),
we commit a Type II error (miss, we do not find an effect when we should have), or we commit a Type II error
(misspecification, we find an effect, but an effect that is opposite of what is actually in the population.
4. Rejecting the null hypothesis does not always guarantee that the research hypothesis is being supported. Some
research hypothesis are not always the opposite of the null hypothesis, the research hypothesis must state the fact
that a relationship is predicted or going to happen or else the research hypothesis is pretty much the null hypothesis.
Therefore, in these circumstances, if one was to reject the null hypothesis than there would be no support for the
research hypothesis because in essence one would be rejecting the research hypothesis as well.
5. First, it is possible to hypothesize that there is a relationship between variables when there really is not. Three
possible relationships predicted could be that there is a positive/negative linear relationship, there is a mean/no mean
difference, and that there is a mean/no mean pattern relationship. Second, two more possible decision errors could
result from retaining the null hypothesis when you should have rejected it. That is saying that you predicted no
relationship but there was a significant relationship. The other error could be from a Type III error, or misspecification.
This says that you predicted a relationship and there was significant relationship, but you predicted the wrong
"direction" of the relationship.
6. Null hypothesis significance testing is what NHST stands for. First you must come up with a research hypothesis. You
will want to determine what the null hypothesis is as well. The research hypothesis can be the null hypothesis. You
will then collect data from a representative sample of the population. You will then run a data analysis and get the
significance statistic and p-value. You decide to reject or retain the null hypothesis based on the p-value. If p is less
than .05 then you reject the null. If p is more than .05 you retain the null and reject your research hypothesis. You can
1.correctly retain the null hypothesis. 2. correctly reject the null hypothesis. With either of these (#1 or 2) It is likely that
you had good measures, a representative sample, and a good design in your study. You could also 3. commit a type I
error - false alarm, you found an effect when really there is not an effect in the population. 4. a type II error- miss - you
did not find an effect but really there is an effect in the population. or 5. a type III error - misspecification - you found
an effect and there is an effect in the population but you found a false one. Type I and III errors can be a result of a
bad design, a nonrepresentative sample, and bad measures. A type II error is likely to result from a bad design,
nonrepresentative sample, bad measures or a sample size that is too small.
7. The Null Hypothesis Significance Testing is the testing of the null hypothesis, the hypothesis of no difference. To test
this, the researcher finds data with respect to a particular RH. This is then tested for significance with a p value. The
various possible outcomes are as follows. One could be correctly rejecting the Null, saying there is something that we
are looking at. Another could be correctly retaining the Null, meaning no meaningful relationship. Yet another could be
a Type I error. This is thinking you found something when there really isn't anything. Bad! A Type II error could also
come up. This is the sad fact that we may miss a relationship when there really is one. Last, A Type III error is a
possibility. This is the realization that there is a relationship, but the direction is misspecified.
8. Null hypothesis significance testing is used to decide whether or not there is a significant differnce in the results of a
study. First, you have to have a research hypothesis. You will either decide in favor of this RH: or the null. They you
must collect data, making sure to control for as many confounds as possible. Third, you perform a statistical analysis
to find out if the results were significant. Finally, you can look at the p value to determine whether or not to retain the
null (retain if p>.05). There are five possilbe outcome for this. First of all, you could have a correct retention, meaning
that there is no difference in the data. This will happen if p>.05 and there is no difference in the population. You could
also have a correct rejection, meaning there is statistical significance. For this to be true, p<.05 and the hypothesis is
true for the population. You could get a Type I error, or a false alarm, meaning you incorrectly thought there was a
relationship where there wasn't one. A Type II error, or a miss, occurs when you think there is no relationship, but in
reality there is one in the population. Finally, you could have a Type III error, or a misspecification. This happens when
you correctly think there is a relationship in the data, but you find the wrong relationship.
9. Null Hypothesis Statistical Tests can result in correct retention or correct rejection of the H0:, or they can result in
Type I, II, or III errors. A Type I error is a "false alarm," where an association that doesn't exist is identified and the null
is falsely rejected, usually due to researcher error. A Type II error is a "miss," in which the researcher flasely retains
the null, usually due to a lack of power. Finally, a Type III error is when the null is correctly rejected, but the
researcher finds an effect in the wrong direction. Type III errors can be caused by incorrect statistical manipulations.
10. NHST stands for Null Hypothesis study testing. The five possible outcomes are correct rejection of the null and this
means that we have found an effect, correct rejection of the null means that we have not found a significant effect.
Pattern differences mean that we have made a mistake along the way and our hypothesis does not correctly match
what we have found in the population. Mean differences mean that we have found an effect but the sample size was
too small so we could not find a significant effect.
11. Null Hypothesis Statistical Testing. This process is basically the key in determining whether or not you can accept or
reject statistical analysis of a research hypothesis and if you need to replicate the data. The five possible outcomes
are: 1. Type I error in which you incorrectly reject the HO: (you find a relationship when there isn't one) it is also called
a False Alarm. 2. Type II error in whcih you incorrectly retain the HO: (you decide there is no relationship when there
is) it is also called a Miss. 3. Type III error in which you correctly state the that variables are related but you get the
direction of the relationship wrong, it is also called a misinterpretation. 4. Correct retention is when you correctly
decide that there is no relationship between the variables. 5. Correct rejection is when you correctly decide that there
is a relationship between the variables.
12. few steps. First it is important to look at the RH and its direction, also look at the HO. Then data needs to be collected,
making sure that the sample is representative of the population. Last is the data analysis which is where your going to
use the bivariate statistical models and use the p-value, etc. The first 2 outcomes are positive outcomes and they
include correctly rejecting the null or correctly retaining the null. These 2 outcomes occur due to good measurements,
good sample, and good design. The last 3 outcomes are all negative outcomes. You could have a Type I, Type II, or
Type III error. Type I and Type III errors occur due to poor measurements, poor sample, and poor design. Type II
error occurs due to poor measurements, poor sample, poor design, but also, poor sample size (not enough in the
sample).
13. The Null Hypothesis Statistical Test is used to determine a relationship between variables. It is used to prove the
direction of variables within a study. If p > .05 we retain the null, if p< .05 we reject the null. The RH can be the HO.
The five possible outcomes are: The first two outcomes are a correct retention or a correct rejection. The next
possible outcome is a type 1 error or false alarm. We think there is a relationship when there really isnt. The next
possible outcome is that there is a misspecification or type III error. There is a relationship but it is different from the
one we specified. Either a type 1 or 3 error can be caused by too small a sample, or bad sampling. The last outcome
is a type II error or miss. This says that there is a relationship but we missed it.
14. The process of NHST, null hypothesis significance testing, is used to determine if there is a statistically significant
relationship between variables. The five possible outcomes are correct retention of the null hypothesis, incorrect
retention of the null hypothesis (called a Type II error or miss), correct rejection of the null hypothesis with the pattern
of relationship in the correct direction, incorrect rejection of the null hypothesis (called a Type I error or false alarm),
and correct rejection of the null hypothesis but with a pattern of relationship opposite to the actual pattern of
relationship in the population (Type III error or misspecification). The likely reason for correctly retaining the null
hypothesis or correctly rejecting the the null hypothesis with the pattern of relationship in the correct direction is that
the sample size is large enough and there are no confounds in the experimental design. The likely reason for a Type I
or III error is that there are confounds in the experimental design. The likely reason for a Type II error is that there are
confounds in the experimental design or the sample size is too small to provide sufficient power.
15. The process of null hypothesis testing(NHST)starts with a research hypothesis. This can either involve a direction or
involve the null hypothesis itself as the research hypothesis. Then you select a data sample that represents the
population you wish to study. You perform the experiment, and then run a data analysis. If the significance test, pvalue, is less then .05, you reject the null hypothesis. If it is larger than .05, you must retain the null hypothesis. This
can result in five different outcomes. One is correct retention of the null hypothesis if the effect is found to not be
significant. The second outcome is a correct rejection if the results are found to be significant. A third outcome is a
Type I error, a false alarm. This is when you get a non-existent effect. A fourth outcome is a Type II error, a miss. This
only occurs when you retain the null hypothesis because even if you found an effect, it was deemed insignificant even
though it was significant. You can statistically determine the possibility of a Type II error by subtracting the power of
the experiment from 1. The final possible outcome is a Type III error, which is also known as misspecification. This
occurs when you find a significant effect, but do not get the correct direction of the effect. Type I and Type III errors
are possible only if you reject the null hypothesis where as Type II is only possible if you retain the null hypothesis.
16. NHST stands for Null Hypothesis Significance Testing. The first step in this process is to make a research hypothesis.
Then you should collect data from a sample that represents the population and analyze the data using F, X2, or r. The
possible outcomes of NHST are to correctly reject or correctly retain the null hypothesis, which both show that the
design of the study was good. You could also commit a type I, type II, or type III error, all which show that the design
was bad and that perhaps the sample size was too small.
17. NHST stands for null hypothesis significance testing. First you form a research hypothesis, which can have a direction
or be the same as the null. Then you gather data that represents the population and begin analyzing that data. You
need to find the summary statistic (r, F, or X2) and then find the p value. If p>.05, you should retain the null
hypothesis. If p<.05 you should reject the null hypothesis. After making your decision, the five outcomes are:
1)correct retention of the H0: 2)correct rejection of the H0: 3)Type I Error (false alarm) You mistakenly reject the null
and say there is an effect when there really is not one in the population. 4)Type II Error (miss) You mistakenly retain
the null and say that there is no effect when there actually is one in the population. Type III Error (misspecification)
You correctly reject the null, but the effect you found is the opposite from the majority of research.
18. Null Hypothesis Statistical Testing is used to analyze bivariate statistics. The process begins with a research
hypothesis. This hypothesis can predict varying relationships, or it can be the null hypothesis (which predicts there is
no relationship). After the RH is decided, data is collected and a statistical model is chosen. After analyzing the data
with the correct statistical model, we find our results. These results include a p-value (significance) and an effect
(either one predicted by the RH or a contrary one). Based on the significance we decide which of the five outcomes
we have found. If the significance is less than .05 we will reject the null hypothesis, determining that there is a
relationship between the variables. If the p-value is greater than .05 we will retain the null hyopthesis (that there is no
relationship between the variables). This leads us to five possible outcomes. The first possibility is that if the null
hypothesis is true for the target population and we find a significance of over .05, then we will correctly retain the null.
Second, if the null hypothesis is false for the target population and we have a significance of less than .05 we will
correctly reject the null. Third, the null hypothesis may be true for the target population, but we may find a significance
of less than .05 (therefore we incorrectly reject the null) leading to a type i error, or a false alarm. Fourth, the null
hypothesis may be false for the target population, but we may find a signifance of greater than .05 (thus incorrectly
retaining the null). This leads us to a type ii error, or a miss. Finally, the null may be false for the target population, and
we may have significance of less than .05 (correctly rejecting the null) but we may find our relationship to be the
opposite of what is true for the target population. This results in a type iii error, or a misspecification. Type 1, and type
3 errors are likely to result from an unrepresentative sample, bad data, or a bad design. Type two errors may result
from these things, but they may also result from a sample size that is too small.
19. It stands for the Null Hypothesis Significance Test. It starts with generating a research hypothesis that either has a
direction the research wants to go, or can also be a null hypothesis. From that point you need to test the RH using a
representative sample from the sample population. After gathering the data it needs to be analyzed by one of the four
tests (r, chi-square, BG/WG ANOVA). After running these tests one five outcomes are possible: 1, correctly rejecting
the null hypothesis by examing the p value and finding valid support; 2, correctly retaining the null hypothesis by
having too large a p value and results are contrary to what you hoped to discover; 3, Type 1 error (false alarm), think
you found an affect that isn't really in the population, could be the result of bad measures and poor sampling; 4, Type
II error (miss), this is only when you retain the null hypothesis, and claim there is no affect when there really is one in
the population. Can be the result of too small a sample or insufficent power. And 5, a Type III error (misspecification)
finding an effect but opposite of that which is really present in the population, caused by poor sampling, and bad
measures.
20. Null Hypothesis testing is done by first getting a summary of the F and df amounts and using the decided statistic.
You must also tell whether the p value is greater of less than .05. If it is greater than .05 you will retain the HO; if it is
less than .05 you will reject the HO. The five possible outcomes are as follows; Correct retention of the HO, which is
when p>.05; correct rejection of the HO, which is when p<.05; type I error, also a false alarm, is when you assume
two variables are related; type II error or a miss, is when you don't assume two variables are related; type III error or a
misspecification, is when you correctly assume the variables were related, but the pattern is wrong.