![DrGreggHarbaugh](/img/default-banner.jpg)
- Видео 117
- Просмотров 82 735
DrGreggHarbaugh
Добавлен 9 янв 2012
The 1.5×IQR Rule To Locate Outliers & Modified Box-&-Whisker Plots
In this short video, we follow on from the last video in which we introduced the box-and-whisker plot. Here, we introduce the 1.5×IQR rule to locate outliers using the 5-number summary. We provide a rationale for the 1.5 multiplier (resulting in an interval of usual values the length of 4 measures of spread for the data set). We also extend the idea of mild and extreme outliers (using z-scores of 2 or 3 in magnitude) to this new rule, and we provide a protocol to classify outliers as mild or extreme using the 1.5×IQR rule. Finally, using a realistic data set, we demonstrate how to modify a box plot to indicate the presence of outliers (along with a variation that allows for the distinctio...
Просмотров: 3 417
Видео
Calculating Percentiles, Quartiles, and the 5-Number Summary Using Excel
Просмотров 1523 года назад
In this very brief video, we demonstrate how to calculate the 5-number summary using Excel (or Google sheets). This introduces the =PERCENTILE(·) function in Excel. We then redo the calculation for the 5-# summary using just the =QUARTILE(·) function.
Quartiles, the 5-Number Summary, and Box-and-Whisker Plots
Просмотров 403 года назад
In this brief video, we introduce the concept of the quartiles for a data set. We provide two protocols to locate these values: the first is the by-hand method of locating the median for the upper- and lower-halves of the data, and the second is using percentiles and the L-locator method. After defining the quartiles and providing a means of calculating them by-hand, we introduce the 5-number s...
Introduction to Percentiles (part 2)
Просмотров 453 года назад
This is the 2nd of a 2-part video in which I introduce percentiles. In this video, we focus on authentic percentile problems, given a specified percent, locate the value in the data set for that percentile. (In the previous video, we were given the value and looking for the percent...this is the reverse of that type of problem.) The by-hand protocol demonstrated here is often called the L-locat...
Introduction to Percentiles (part 1)
Просмотров 1113 года назад
This is the 1st of a 2-part video in which I introduce percentiles. In this video, we will briefly explore the connection of the median to percentiles. We will start with our working definition for percentiles (the value in the data set for which at least k% of the data are at most the percentile). In the process of introducing this new concept, we will introduce a variation on index notation t...
Introducing the Protocol To Calculate the Median (by-hand)
Просмотров 273 года назад
In this short video, I present the protocol to calculate the median of a data set. Along the way, I show how to use the =SMALL(·) & =ROW(·) functions to sort a data set into one column in Excel (or Google Sheets). Using two small data set (sets of odd prime numbers), I demonstrate the different protocol to find the median for a data set with an odd number of data values compared to a data set w...
Statistics...1000 ways to do the same thing: Another a 2nd approach to summarize quantitative data
Просмотров 263 года назад
In this brief video, I discuss a troubling element of statistical analysis and data science that "haunts" many students. That is the idea that there appear to be 1,000 ways to do the same thing...and this happens again and again in statistics. In truth, this is a fairly accurate observation. While it does cause issues for students learning the discipline, it actually has some potential benefits...
Demonstration Problems Working with the Count-by-Thirds Empirical Rule
Просмотров 813 года назад
This brief video walks through a handful of estimation problems using the count-by-thirds empirical rule. In particular, we explore problems that use the intervals associated with the rule, intervals with one end-point at the mean, and intervals that require a combination of intervals. Next, we will explore problems that inquire about estimating the upper and lower percentiles of normally shape...
How the Standard Deviation Changes with Additional Values in a Data Set
Просмотров 6 тыс.3 года назад
This brief video explores how the mean and standard deviation changes when additional data is added to a data set. In particular, we explore what happens to the mean and standard deviation when a new data value that equals the mean is added to the data set. And then we explore what happens to the mean and standard deviation when two new data points are added that are 1 standard deviation above ...
Introducing Statistical Bias: Or...¿why are there two formulas for the standard deviation?
Просмотров 2563 года назад
In this brief video, we introduce the concept of statistical bias. First, we start by showing empirically that the mean of a sample is unbiased (i.e., is good at estimating the population mean, in the long run), but the maximum of a sample is biased. Next, we examine how sample data produces a biased estimate for the variance of a data set when using the population formula for variance, but thi...
Excel Functions for Mean and Other Summary Statistics
Просмотров 113 года назад
In this brief video, we review how to find the mean and standard deviation of a small data set using Excel (or Google Sheets). The array notation along with the column (and row) selection notation are presented. The functions demonstrated here include =COUNT(·) (for sample size), =MIN(·) & =MAX(·) for the minimum and maximum, and the =VAR(·) function for the (sample) variance.
The Count-by-Thirds Empirical Rule...and truly useful and pragmatic rule of thumb for normal data.
Просмотров 1493 года назад
In this brief video, we introduce the count-by-thirds empirical rule. This is a more refined variation of the traditional empirical rule that counts by thirds of a standard deviation (instead of whole standard deviations) and utilizes far more pragmatic and useful percentages for approximation purposes.
The 68-95-99.7 Empirical Rule...and why it is a less-than-desirable tool.
Просмотров 513 года назад
In this brief video, we introduce the conventional empirical...often called the 68-95-99.7 (or worse, the 68-95-99) rule. The "accuracy" of the rule is demonstrated with a nearly normal data set. The video ends with a critique on why this is not a useful or pragmatic rule to use (in its given form).
Revisiting Measurement Levels
Просмотров 123 года назад
This brief video discusses the key distinctions between nominal, ordinal, interval and ratio level data. There is a detailed demonstration to elaborate the difference between ordinal and interval data. There also is a discussion of the difference between interval and ratio level data. Additionally, the limitations and uses of each level of data are discussed and compared.
Exploring Grouped Frequency Distribution Tables: Common distribution shapes in GFDTs and histograms
Просмотров 713 года назад
This brief video is part 7 of 7 in a series about summarizing and visualizing quantitative data using GFDTs and histograms. In this video, we present some of the common shapes for distributions you are likely to encounter when working with data.
Exploring Grouped Frequency Distribution Tables: Picking a "nice" class width for a GFDT.
Просмотров 2273 года назад
Exploring Grouped Frequency Distribution Tables: Picking a "nice" class width for a GFDT.
Exploring Grouped Frequency Distribution Tables: Using the flowchart to build a GFDT by hand.
Просмотров 613 года назад
Exploring Grouped Frequency Distribution Tables: Using the flowchart to build a GFDT by hand.
Exploring Grouped Frequency Distribution Tables: The rules for building GFDTs.
Просмотров 523 года назад
Exploring Grouped Frequency Distribution Tables: The rules for building GFDTs.
Exploring Grouped Frequency Distribution Tables: GFDTs with very large class widths.
Просмотров 163 года назад
Exploring Grouped Frequency Distribution Tables: GFDTs with very large class widths.
Exploring Grouped Frequency Distribution Tables: The impact of the 1st lower class limit on a GFDT.
Просмотров 123 года назад
Exploring Grouped Frequency Distribution Tables: The impact of the 1st lower class limit on a GFDT.
Exploring Grouped Frequency Distribution Tables: The impact of the class width for the GFDT.
Просмотров 1213 года назад
Exploring Grouped Frequency Distribution Tables: The impact of the class width for the GFDT.
Exploring Grouped Frequency Distribution Tables: The difference between FDTs
Просмотров 273 года назад
Exploring Grouped Frequency Distribution Tables: The difference between FDTs
Measuring Spread for Bell-Shaped Distributions: Setting the Stage for the Standard Deviation
Просмотров 923 года назад
Measuring Spread for Bell-Shaped Distributions: Setting the Stage for the Standard Deviation
Grouped Frequency Distribution Tables (GFDTs): Visualizing & Summarizing Quantitative Data
Просмотров 7683 года назад
Grouped Frequency Distribution Tables (GFDTs): Visualizing & Summarizing Quantitative Data
Stem-and-Leaf Plots: Visualizing & Summarizing Quantitative Data
Просмотров 9743 года назад
Stem-and-Leaf Plots: Visualizing & Summarizing Quantitative Data
Tally Plots, FDTs & Frequency Polygons: Visualizing & Summarizing Quantitative Data
Просмотров 1393 года назад
Tally Plots, FDTs & Frequency Polygons: Visualizing & Summarizing Quantitative Data
Expected Value for the Negative Binomial Distribution
Просмотров 11 тыс.3 года назад
Expected Value for the Negative Binomial Distribution
How to Build a GFDT (for Discrete Data) Using Excel's =COUNTIF(·) Function
Просмотров 823 года назад
How to Build a GFDT (for Discrete Data) Using Excel's =COUNTIF(·) Function
Using Excel to Conduct a Chi-Square Goodness-of-Fit Test (4/4)
Просмотров 364 года назад
Using Excel to Conduct a Chi-Square Goodness-of-Fit Test (4/4)
Using Excel to Conduct a 3x2 Chi-Square Test of Independence (3/4)
Просмотров 1,2 тыс.4 года назад
Using Excel to Conduct a 3x2 Chi-Square Test of Independence (3/4)
Nice work, thanks!
Thank you so much for this video. This saved me!
great video, I think you meant stacked bar chart and not stacked pie chart?
Good illustration, but I couldn't stop wondering (and being sceptical) - in relation to the probit model - of the statement early on that since we CAN'T observe y* (the latent variable), we can simply ASSUME it has a normal distribution. What on earth is the theoretical justification for this? Presumably a similar assumption is implied by the logistic regression's standard functional form (i.e. a linear relationship between the log of the odds ratio (the dependent variable) and the explanatory variables, though the video is completely silent about what the theoretical justificationbof that implied assumption might be...
A really great explanation! Thank you! Some minor mistake in writing: at 14:19 you wrote F_Unlikely = exp(-0.82) but you meant 0.44/(1+0.44) instead. Same for F_somewhat Thanks!
thank you
Honestly, this is exactly what I needed for a refresher for stats work at work. I especially appreciate that you took the time to explain the motivation of IQR. Thank you so much, Dr. Harbaugh
thank you for this helpful video Dr Gregg
Thank you very much.
Wonderful explanation 🥰
Thanks very muuuuuuuch. I wish I had found this video sooner.
This is an amazing tutorial. Thank you very much!
It was very useful. Thank you!
This video sucks. It assumes that the mean doesn't change when a new data point is added. I'm looking for the effect on both mean and standard deviation when the new value doesn't give the same mean and standard deviation as the original sample.
chill dude 😂
@@cece8309 noo, I needed this for my finals. This man almost made me fail my exam ☹️
Thank you so much for this useful video! Answered all my questions
dear can you help me i work my study contain likert scale for both dependent and independent variables then how can i analysis using ordinary regression analysis was it possible to transform thier response into other means or sum then use ORA
Amazing video, thanks!
This was an amazing video
Wonderful video, thank you! One question, how would we be able to state the effect of an independent variable on the outcome? For instance, how can I express whether or not public school attendance is significantly associated with an increase likelihood of applying to grad school?
I believe a p-value of less than 0.05 for the predictor is considered significant. I'm not sure how to get p-values in ordinal regression though in R. There may be a different metric for measuring variable significance in ordinal regression.
Thank you so much! This is extremely helpful. I do have one question - at 14:20 shouldn't the probability equations be F(unlikely) = 0.44/(1+0.44) = 0.306 and F(somewhat) = 3.58/(1+3.58) = 0.782 instead of exp()?
Really very interesting lecture
THANK YOU SO MUCH!!!! IT WAS REALLY HELPFUL :)
This is the most helpful video talking about the intuitive concept behind the ordered regression that I can found on RUclips. Thank you so much.
9:52
Thanks for this useful video. Can you please tell me how we should deal with these outliers (either mild or extreme ones)? How we should modify it? there is a method called winsorization in which we replace the outlier by the 95th percentile value. For my dataset, when I do that it has some contradictions with the IQR rule.
Thanks! this is very helpful! I'm wondering if there is a way to assess the effect of a predictor X on the thresholds (intercepts) estimation. Are the thresholds assumed to be constant and pre-determined?
Thanks
Thanks for this easy but informativ tutorial! It would be nice if you can do a tutorial on hierarchical ordinal regression! & is there a heuristic on how much categories there sould be at maximum?
The video could have better explained how to arrive at the regression coefficients and intercept when dependent variables have dummy values.
Nice video. thanks for sharing. is there a way to proof that the expected value for a negative binomial is equal to 1-p
This is great. Thank you.
How did you calculate the probability values for the ordinal probit regression model? For the instance with the specific student, how did you calculate the probability of them answering "somewhat" as 0.449?
Thanks for a helpful video. I'm new to ordinal regression and your's was the first video I've watched. The point where I was lost was the animation of bell curve movement with changes in GPA. How can the bell curve move if the value of boundaries is already defined? I can see how we'll get different Y* value with changes in GPA. However, I can't see how bell curve itself will move around on a likert scale!
The y* value comes from the equation, and the y* is the location of the center (mean) of the bell curve. So, say x increases, then y* increases, and thus the center of the bell curve increases along the axis. I hope this helps.
Wonderful, love it!
Great explanation of the graphs
Amazing introduction and example! Thanks a lot!
Many thanks sir
Thank you very much sir ☺️
Love it, you solved a lot of mysteries in my head. :)
Is the effect size in this example Cramer's V?
Excellent video . Thank you.
In the Probit Regression, how did you get the specific values for the tau cuts?
Thanks for the lesson! Is it possible to provide the dataset you have used here?
How do I know which one (probit or logit) fits my data better?
Amazing walk through!!!!! Thank you so much
Thanks for this
this is an excellent video!! khan academy level quality :) explained a difficult concept in a way that was easy for me to understand and visualize
Thank you.
great tutorial. loved it!
Thanks for the great explanation! When I tried to calculate an odds ratio I only got one value. I have trouble interpreting this value. Do you have any advice on that?