Contains lectures related to statistical techniques and theory. Also contains short screencasts demonstrating SPSS, Excel & Intellectus analysis techniques.
I can try. What are you attempting to do with your data? What kind of data? What your outcomes? What kinds of questions are you asking (e.g. descriptive, correlational, group comparison)?
@@tacappaert Thank you so much for offering to help! I am using the Scottish Election Survey dataset from 2021, testing 3 Hypotheses. I am struggling to find which test to use when looking at my hypotheses of "Social media users are more likely to politically participate than non-users". I am finding myself struggling with the variables I am using currently. There is a variable regarding if the respondent is active on twitter, and another for active on facebook. My intention was to recode this into one variable and call it social media user/not. Then for participation, I am using a variable which is a scale from 1-10 on how likely to vote. The university has barely taught us how to use SPSS so I began looking to your videos for help as I feel incredibly underprepared at the minute, any help at all would be brilliant!
Hi, I need your urgent assistance. I wish to perform the analysis of a data set to see whether the patients have improved after the treatment was started. How should I proceed? Thanks in advance
If you have a numeric outcome measure before treatment was started and then will remeasure that outcome after completion of the treatment, i would recommend a dependent samples t-test.
Apparently whoever wrote the Excel code for histograms didn't bother testing it. I'm seeing the same problem in your frequency table and chart as in mine. Everything is shifted incorrectly. Both the table and histogram show a data point in the 90-95 range even though the highest data value is 87. All the data is shifted, though, not just that one. DOY! As far as I can tell, there is no way to set a range for the bins to correct for this absurdity.
In GLM, you can use the button 'plots', then for horizontal axis you would choose 'sex', and for separate lines you would choose 'type of class', and then press the button 'add'
Good video and information. I saw that the numbers or data set is not in ascending order; is not matter to get the quartiles results? Thank you for teaching.
This is exactly what I am looking for to complete my thesis ... thank you so much ... I trued the syntax that you copied in the description box and I keep getting errors ... can you help me with that ?
Hi thanks for the video. I have two variables with 6% and 7% missing. The Little's MCAR test is insignificant. I have missing values on categorical data as well as other variables which are measured with Likert. The sample N=299. I am a newby and wondering if I can use EM imputation for both categoricals and others.
Great video! I have a question. I am conducting research on the effect of virtual learning on long-term self-esteem. I am looking at whether their camera was ON or OFF during the duration of the survey, age, gender, and whether students were online/in-person during the school year. I am looking at those factors in terms of scores on the Rosenberg Self-Esteem and the Multidimensional Body Self Relations Questionnaire (with 5 different subscores within the survey). All scores are using valued Likert Scales. Since I have multiple independent and dependent variables, would a two-way MANOVA or a Canonical correlation be appropriate to use? Thanks.
Hi Sarah, glad you found it useful. Yes, if you have multiple independent variables and multiple dependent variables that have a statistical relationship then a two-way MANOVA would be appropriate.
It gives different F and significance values when simple effects analysis is done way. Split file command option also gives the same results. But the syntax given by Andy Field (2013) gives different F and p values which are more conservative than the ones obtained through the above procedure. Can you help me understand the reasons for this discrepancy? Here's Andy's code for reference where IV-1 and IV-2 to be replaced with variable names: MANOVA Attractiveness BY IV-1 (0 1) IV-2 (1 3) /DESIGN = IV-1 WITHIN IV-2(1) IV-1 WITHIN IV-2 (2) IV-1 WITHIN IV-3 (3)
Hi Brad, in respect to the analyses and the outputs/results you get, there should be no difference. The interface and the menus in Intellectus are far easier to navigate in my opinion.