Psych#Stats in Sydney, Australia | Education

Psych#Stats

Locality: Sydney, Australia

Phone: +61 469 765 964

Address: Bella Vista Sydney, NSW, Australia

Website:

Likes: 110

Add/Update information

Reviews

Tags

• Businesses
• Education
• Tutor/teacher

17.01.2022 Which Statistical Test???? Always confusing, isn't it? Most of the time the answer lies within the question! For example, words like association, predicted, difference, controlled for, moderated, additional variance explained. There are certain key words we should be looking for... (Word) -- (Test) Association -- Correlation Predict -- Regression Controlled for -- Multiple regression Additional variance explained -- Hierarchical regression Difference -- t Test/ANOVA ( of some kind) https://www.facebook.com/PsychStats.Au/

17.01.2022 No Entry, One-Way, Two-Way etc. etc ANOVA Easiest way to get our heads around ANOVA is to get some kind of picture in our head. Maybe different colors or patterns representing the ever confusing this way, that way, one way, four way, this effect that effect I am attaching a quick reference page attempting to offer you a line sketch of that colorful picture and hope You get the picture! This is based on fact that essentially ANOVA is a ‘cell design’. This means that w...e take information on bunch of people who are almost alike in all aspects except their names are different and put all that information in one cell. I just explained the jargon- Group Mean Anyway, the page builds on this ‘cell’ and shows some elementary ‘Between subjects ANOVA’ & main effects. At the end there is an example of ‘Within Subjects ANOVA’. The purpose is to demonstrate Why we require less participants in within subjects design? SPSS, Stata or any other software on which we can run ANOVA, compares these "cells" and tells us how likely its is ( AKA: p value) for all these cells to be nearly equal. If that p value ( or Sig. as in SPSS) is below 0.05, we say it's significant! Any questions , please contact https://www.facebook.com/PsychStats.Au See more

16.01.2022 p Value, Effect Size etc.: Lets think of Michael Jordan's height. We can then consider effect size (for simplicity) as - "The difference between MJ's height and the average height of male population in USA". Now let's think of a chance of finding people of MJ's height in a crowd of 100 waiting at New York subway platform. How many we would find - 4 or less or unlikely?.. There! That’s our 'p value' (number of people with MJs height/100)!! Next, let's take this a step furthe...r and look at significance value.... Somehow, if we find 6 or more people out of those 100, who are as tall as MJ (that's p>!!), we will need to conclude " there's insufficient evidence to reject null hypothesis". All because there were 6 or more really tall persons on the platform at that time! Do you think this happened because, we looked at just one platform. And maybe if we look at more platforms or may be if we consider a larger crowd, we will almost certainly find less than 5 per 100 people as tall as MJ? That’s us saying ‘the experiment didn’t have enough power’ or this is a type II error!! Easy? https://www.facebook.com/PsychStats.Au/ See more

16.01.2022 I don’t understand statistics I am scared of statistics I HATE statistics These are the 3 most common grievances of psychology and nursing students. And frankly, its not their fault! If a statistician approaches a random person at Cole’s and asks about ‘Mean Positive Affect’ or ‘Risk Ratio of Cancer’, what do you think the person will say? First and foremost, there is nothing to worry, the actual ‘Statistics’ are churned out by SPSS, Stata, Excel, or some other software ...Continue reading

14.01.2022 Taming the monster -- "Standard Deviation"! What is it? How to calculate it? Say on 1st June a nice coffee place has reopened after the shutdown. Three friends decide to meet up there and have coffee (after nearly 3 months!). As luck would have it, EFTPOS machine is not working. Cash Only!... They dig up the coins from pocket and come up with $3, $4 and $5. Large coffee is $4 so they get 3 large coffees. While they are enjoying their coffee, let’s look at what happened. All of them did NOT have equal amount of money (AKA: Variation in sample). However, being friends, they pooled together what they had (AKA: Sum of variable) and got 3 coffees of equal price (AKA: Mean was calculated and used to represent the sample!). Now comes the fun part! Did all of them have exactly $4 . No, right? So, let’s try to see how much money, on an average did these 3 blokes were away, from the magic figure of 4. Ok, so first one was 1 dollar short, one had exact amount and the last one had a dollar extra, right (AKA: Difference from mean)? In numbers it would be -1, 0; if we add up to get the total difference, it’s 0? So, the average difference is also 0?? Wait! What just happened?? We calculated difference from mean, then we added those differences and tried to calculate the average difference. Think about it. this is bound to happen in any sample if we just add up the differences. Some differences will be negative, some positive and the total will always be zero! To overcome this issue, each difference is multiplied by itself (AKA: Squared). Then all those ‘squared’ differences are added (AKA: Sum of Squares/ Total Variation). Now to bring it back to per friend difference from the average coffee price we just divide it by the number of friends and call it Standard Deviation!!!! I know, I know some of how would say, we are asked to divide the sum of squares by (n-1), but this post is just to give you an idea of the basic steps. I will be doing separate post for degrees of freedom in which this(n-1) will be explained. FOR second question HOW? Answer is ‘Don’t worry’ J because SPSS, Stata or Excel will do it for. https://www.facebook.com/PsychStats.Au/

06.01.2022 Regression is just the next step of what we all, well.. most of us, learnt in school mathematics, unless maths was dropped. It's just step ahead from "Rise over run". What was plotted then on X axis is now the 'predictor' and Y axis now represents "outcome/dependent variable". We used to draw the line through points.... Intercept in the SPSS/STATA output is just the number where the line would cut through Y axis. The scary monster of "Unstandardised Beta" for predictor is si...mply "Rise over run" which we calculated few years ago by hand!!! We can think of simple regression as plotting a line on graph paper when we are given a few "pairs" of numbers. One difference probably is, then we thought of pairs of numbers and today we split the same pair into an 'independent variable' & a 'dependent variable'. We can think of multiple regression with two 'independent variables" or predictors as trying to plot a 3D graph. Imagine any table with two predictors as axes X & Z starting from any corner (you like ) and the dependent as Y axis standing vertically up from the same corner.. Hope you get the picture!! ( https://www.facebook.com/PsychStats.Au/ )

05.01.2022 Statistics = Rocket science Yes / No

05.01.2022 Horrible Homoscedasticity & Lame Linearity! Whether it’s a simple Pearson’s correlation or regression analysis, the assumptions of homoscedasticity (or homosKedasticity as some sources refer to it) and linearity are very important. Let’s understand what it is and why is it important in plain-speak. Say we are trying to predict (AKA: Regression) or associate (AKA: Correlation) a person’s height based on weight. The regression equation or the fit line in scatterplot will idea...lly travel through say we think of scatter plot as a road then it would be the white line running from middle of the road! Linearity simply means that the road should be straight, it should not be C, U or S shaped, that’s it! For the sake of simplicity, homoscedasticity means that the width of road should be same from beginning to end! You can think of beginning of the road as lower values of predictor / standardised predictor or any thing that you choose to have on X axis. The higher values would mean the end of the road. All that meeting this assumption means is that the width of road is same (sort of) through its length, most importantly it’s NOT tapering at either ends. Why are these important Well, lack of linearity and presence of a shape (C,S or U) would mean that correlation/linear regression is not the correct way of representing the relationship. We need to look at alternate tests. And a lack of homoscedasticity would mean that the our estimates from correlation and regression will be ‘biased’. https://www.facebook.com/PsychStats.Au/

03.01.2022 Free 1on1 Zoom Session on Moderation Analysis I have recieved a quite a few enquiries for interpretation of Moderation Analysis using PROCESS. I will be running 6 free of charge zoom sessions of 30 minutes each between 6 to 6.30pm & 9 to 9.30pm starting from Monday 21 September to 23 September. Please message here.

01.01.2022 Just a quick note on ANCOVA. Before running the actual ANCOVA, besides the routine assumptions like normality etc; there’s an important ANCOVA assumption of ‘Independence of Covariate’. In case this is ignored and IF the covariate is not independent of the independent variable (IV: treatment levels/groups) then the final result could be either suppressed or inflated. Best way of managing this, is to test if the covariate differs across the levels of IV. If the IV has only two levels (male/female; Australian/non-Australian etc.), an independent samples t test with covariate as dependent should be non-significant. In case of more than 2 levels of IV, a simple one way ANOVA with covariate should be non significant. Either of these would reveal that the covariate is related to the IV! Assumption MET!!!