Can't agree more ,don't know why they like to make everything harder than it is I used to hate these statistical things but after searching and learning from an outsource i really liked it and i started thinking about working as a data analyst
I am currently having to do some T-Test analysis as part of my MSc. For someone who has never encountered statistics before (at least not for a while), you have really helped me to understand this process. Many thanks.
Amazing visuals, but please stop switching between examples... First it's patients and blood pressure, then it's chcolate and grams, then patients and pain relief...
Great video! My only critique/suggestion might be that it would also have been nice to see the actual math. That might be beyond where you wanted to go with this video, so maybe it’s a “part 2” or “Advanced” follow on video ;-). Anyhow, thanks!
This isn't valid. The most important part of p-value is that its threshold must be defined before the study. And typically the threshold 0.05 is chosen, but *why* ? The typical threshold of 0.05 is completely arbitrary, unscientific, and it means of 100 studies we would expect 5 with the wrong result. It doesn't mean the result is statistically significant, it means the result is considered statistically significant, even if it really isn't.
Thank you so much for the very good and useful presentations I really loved it !!! Can we use these slides in our presentations? or they are copyrighted materials?
I'm a grade 10 high school student from the Philippines and we had to make an research. We need to use the T test in our research thank you so much I learned so mich🙏
Hi, I have 10 Pairs of Land Surface Temperature (LST) data from 10 sites for two years. If I want to conclude that overall T-Test suggests there are significant difference between the two years, do all 10 pairs need to be significant individually? or if majority sites show significant difference, I can conclude that the overall result is significant?
Hi : ), when you're analyzing data from 10 different sites for two years and aiming to use a t-test to determine if there's a significant difference in the Land Surface Temperature (LST) between the two years, it's important to understand the nuances of statistical significance and how it applies to your dataset. Whether each of the 10 pairs must individually show significance or if a majority is enough to conclude an overall significant difference depends on your analysis approach and how you structure your hypothesis testing. Here are some key points to consider: Individual vs. Combined Analysis: If you're conducting individual t-tests for each of the 10 sites, then each test assesses the significance of the difference in LST between the two years for that specific site. A significant result in this context only applies to the site tested. You cannot automatically infer an overall significant difference for all sites unless you're aggregating the data or results in a statistically valid way. Aggregate Data Analysis: An alternative approach might be to aggregate the data from all sites and perform a single t-test to compare the years. This method would give you an overall picture but would also assume that all sites are comparable or have been appropriately normalized to account for site-specific differences. This aggregated analysis would allow you to make a generalized conclusion about the difference in LST between the two years across all sites. Majority of Sites Showing Significance: Simply having a majority of sites showing a significant difference does not automatically validate an overall conclusion of significance across all sites, especially if each site is analyzed independently. The variation in significance across sites could be due to various factors, including environmental, geographical, or methodological differences. Meta-analysis or Combined Statistical Approach: If individual analyses are preferred, you might consider a meta-analytical approach or a combined statistical test that takes into account the results from the individual tests. This can help you draw a more comprehensive conclusion that accounts for the variability and significance observed across different sites. Adjusting for Multiple Comparisons: When performing multiple t-tests, one for each site, you're increasing the risk of Type I error (false positives). To account for this, you might need to adjust your significance threshold using methods like the Bonferroni correction, which involves dividing your alpha level by the number of tests performed, or other corrections designed for multiple comparisons. In conclusion, to assert that there's an overall significant difference in LST between the two years based on your data from 10 sites, you need to carefully consider how you structure your analysis. If your goal is to draw a conclusion about all sites collectively, consider aggregating the data for a single analysis or employing a statistical method that can combine the results from individual site analyses. Always be mindful of the implications of multiple testing and the need for appropriate adjustments to your significance criteria. Regards Hannah
Is it possible to do a t test on football stats like goals and assists based on certain factors like women and men world Cup or something like football after and during covid
Hmm, that would have to be looked at more closely, but the question of whether there is a difference between the number of goals scored in a match by men and women can already be tested with a t-test. You only have to check whether the data are normally distributed! If not, you could use the Mann-Whitney U-test. Please have a look here: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-LcxB56PzylA.html
@@blessedmusakaruka1235 If I understand it correctly, you would use a Chi2 Test in this case, please have a look here: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-EjtXk-yEK6w.html
Hello. I just want a simple suggestion as to what type of statistics (Anova, chi-square, etc.) I should use to determine my goal below. I am truly confused yet optimistic for someone generous out there, and I would greatly appreciate any additional comments or suggestions to clarify or simplify my statement or claim here. Thank you in advance. "Among the randomly selected senior students in the eight (4 public and 4 private) US south-west states (California, Arizona, New Mexico, and Texas), their responses are unanimous (strongly agree, agree, neutral, disagree, strongly disagree) as regards participating in home gardening rather than school gardening."
We deal with population groups and of recent are also combining qualitative and quantitative data analysis in our work. The t-Test course has come in handy for us because we refreshed our minds about many of the principles. Thanks.
