answersLogoWhite

0

The answer will depend on the hypothesis that you are testing and the associated model. You could be looking at how many seed from various packets germinate and if the number of seeds in each packet is small, you could expect the numbers to follow a binomial distribution for which exact probabilities can be calculated. If the number of seeds is large then the distribution is more likely to be normal or lognormal. Or, if you are looking at how long it takes the seed to germinate, you may want to use the exponential or the Poisson distribution. If you want to study how germination is affected by soil quality (fertilizer) and watering you may want to use a Greco-Latin design and use a chi-square test. So, you see, you need to put some thought into this.

User Avatar

Wiki User

7y ago

What else can I help you with?

Related Questions

What statistical test do you use for nominal by ordinal data?

Kruskal-Wallis H test.


Is there a statistical test to assess the difference between two ratios. I have numbers which are ratios and want to test whether there is a statistical difference between them?

You can use the z test for two proportions. The link below will do this test for you.


When is it appropriate to use the chi-square test in statistical analysis?

The chi-square test is appropriate to use in statistical analysis when you want to determine if there is a significant association between two categorical variables.


When do you use goodness of fit in statistical analysis?

statistical goodness of fit test used for categorical data to test if a sample of data came from a population with a specific distribution. It can be applied for discrete distributions.


What does representative mean in statistics?

You know nothing about how to use statistical analysis to verify or test validity, do u.


What would you use a parametric test for?

* Always when the assumptions for the specific test (as there are many parametric tests) are fulfilled. * When you want to say something about a statistical parameter.


How can one test for heteroskedasticity in a statistical analysis?

One way to test for heteroskedasticity in a statistical analysis is to use the Breusch-Pagan test or the White test. These tests examine the relationship between the error terms and the independent variables in a regression model to determine if the variance of the errors is constant. If the test results show that the variance is not constant, it indicates the presence of heteroskedasticity.


What is statistical test of hypothesis?

A statistical hypothesis test will usually be performed by inductively comparing results of experiments or observations. The number or amount of comparisons will generally dictate the statistical test to use. The researcher is basically making a statement and assuming that it is either correct (the hypothesis - H1) or assuming that it is incorrect (the null hypothesis - H0) and testing that assumption within a predetermined significance level - the alpha.


What is the use of statistical inference in technology?

What is the use of statistical inference in technology?


Which statistical test do you use if you need a correlation for many pairs such as if you have data on 50 couples and want to know if they are similar or different based on a range of values?

student's t test


Use of statistical techniques in capital market?

Use of statistical techniques in capital market?"


What are the key differences between a chi-squared test and a t-test, and how do you decide which one to use in your statistical analysis?

The key difference between a chi-squared test and a t-test is the type of data they are used for. A chi-squared test is used for categorical data, while a t-test is used for continuous data. To decide which test to use in your statistical analysis, you need to consider the type of data you have and the research question you are trying to answer. If you are comparing means between two groups, a t-test is appropriate. If you are examining the relationship between two categorical variables, a chi-squared test is more suitable.