Amazing Video David. I learned a trick while doing this kind of data processing. Instead of dragging you can just double click when the small plus sign appears as it did during your participant number section. It makes it so much easier.
SPSS has many more sophisticated tools and ways of data cleaning and prep. If you want to keep it as simple as possible, export the data into an Excel file and clean it up as described in the video, and then reimport it into an SPSS file. Otherwise you can Google data clean up and preparation in SPSS for the specific functions and methods.
David Dunaetz thank you so much, I was planing to do that. Another question: I am doing a study about “The impact of fan influencers on fans purchase intention” And I have like 5 independent variables which are trustworthiness, authenticity..etc And 1 dependent variable: purchase intention So to analyse the data, I was thinking to do multiple regression and reliability analyses. So are these two enough to answer my question? Regarding testing hypotheses do I do it before or after the analyses, or is it part of the analyses? And can I do it in excel or SPSS? Thanks
When an item allows for multiple responses, it gets much more complicated. Each possible response needs to be coded as a binary dummy variable. For example, if you allow multiple responses for race/ethnicity (which I believe you should), you might have 5 categories: White, Black, Hispanic, Asian, other/prefer not to state. You need to treat this 5 variables if you want to test for relationships and differences: white vs non-white, black vs non-black, etc. Google "dummy variables" for more information.
@@DavidDunaetz hi David thanks for the answer, I've treated all responses as a dummy variabes. My questions were more about how to analyze this further, can we crosstab it or make pivot table out of it?
This video might be useful: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-_vfzjFJg7U8.html . A lot depends on what your data is measuring and what your hypotheses are.
omg! lol yes i literally burst out laughing when you said i"I should have done that at the beginning " yes it at the beginning! lol because loosing data sucks! its so tedious! hahahahaahahahahahahahahahhahaha
I conducted a teacher's survey on google form, the students selected the teachers name and then rated 19 questions about classroom management as excellent(4), Very Good(3) etc. I want to get the number excellent, very good, etc of each question against the teacher. Please how can I go about it?
It sounds like you're interested in "pivot tables." There are several good videos on how to use pivot tables in Excel on RU-vid which you can easily find. They're a bit complicated, so might find that it's easier to count them directly.
There are two reasons. First, often we need to combine questions to get the variable we're interested in. So if I have nine questions on extraversion, I want to label them so that they are easy to identify when I create a new column for the composite variable. Secondly, I will want to make tables of descriptive statistics, correlations, multiple regression, etc.; the column headings will appear in these tables when using the Data Analysis Tool Pak or pivot tables, so I want to have short, clear column headings.
The specific analysis depends on what you're looking for. For example, if you're looking for descriptive statistics or a table of correlations, Google my name and what you're trying to do. There's a good chance I've made a video for it. For the next step in preparing data for analysis, this video might be useful: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-_vfzjFJg7U8.html .
In general, bar charts are superior to pie charts. If you're using Google Forms, you can see that Google creates them for you under "responses." On Excel, it takes a bit of work. Here's a video: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-PeSynxkJ-UI.html
It's possible, but it's a bit tricky and often not useful. To do so, you need need treat each checkbox as a separate variable. For example, if you gave four choices for a question on ethnicity (White, Black, Asian, Latino), you'd have four variables. The first would be White with two possible values: White and not-White, and so on for the next three. You can then calculate the correlation between the variable White and whatever else you've measured (you could equally do a t-test to compare the difference between Whites and non-Whites).
Here are a couple more videos that follow: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Pns9UhRH9ro.html ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Pns9UhRH9ro.html You can go to the homepage of my channel and find many other videos on stats using Excel.
Hey David! I have a data set of respondents who had participated on my questionnaire survey. The questionnaire had drop down options to choose from. Only one response is valid. In this way, I have collected 228 responses. Now how do I code it on Excel. The video here essentially shows a strategy to be followed when a Likert Scale data is available. What about coding for drop down choice options?
Hi Kamonasish. It demands on what type of data it is. If it's nominal (categorical) with more than 2 choices, it's difficult to analyze quantitatively. However, if it's ordinal, interval, or ratio data, numbers will generally work.
@@DavidDunaetz It is a nominal data. I trying to code each option by assigning them a number and then compute. But the issue is with Pivot tables where the values are not correctly computed and represented.
it's funny when you say 228 responses. I had to prepare a questionnaire project with 17,0000 responders for 50 questions. Whether is 228 or 1M, depending how it's set up, it will be a breeze.
I already have everything into numbers form 1 to 5 1 being Strongly Disagree and 5 Strongly agree! doe sit mean i dont need to changed them to numbers?
+alice p You are correct, if they are already numbers you don't need to change them, unless they are reverse coded. In that case, look up my video on how to unreverse reverse scored items.
@@DavidDunaetz hi, thank you for responding! you have been a great help! i dropped mu stat class recently now i'm lost ! i would anyway as it was hard for me. so now i have 2 psych classes that require a research study and i'm stuck! i have ti all now i don't know how to get the info to write my report as i surveyed 80 students in financial aid delays and student success.
I've added the link to the video description. You can download load it here: www.researchgate.net/profile/David_Dunaetz/publication/346345948_Data_Collection_Exercise_Codebook/links/5fbe99b5299bf104cf796aa9/Data-Collection-Exercise-Codebook
ok wheres the next video? now what? i suck at stats! i need to type my report for Monday and im stuck!!!! i need average of women wHo said what and man and standard deviation etc! HELP ME!!!!!!! SOMEONE ! ANYONE!!!!!!! LOL
+ Alice P For Windows: How to Calculate an Average: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-dZk03_tBTO0.html How to Calculate a Standard Deviation: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-mLUn-F6iZ8c.html If you need help on a Mac, you can google around to find the Mac version of the same videos.