1 00:00:00,000 --> 00:00:02,005 Welcome to Purify Data Pro. 2 00:00:02,255 --> 00:00:08,357 Let's walk through your first data screening session using the app's current interface and built-in example. 3 00:00:08,607 --> 00:00:10,697 You do not need an account or password. 4 00:00:10,947 --> 00:00:14,019 Statistical calculations run in your browser. 5 00:00:14,269 --> 00:00:18,792 Choose Open workspace for your own data, or Try sample data to practise. 6 00:00:19,042 --> 00:00:21,133 For help, open the three-dot menu. 7 00:00:21,383 --> 00:00:24,967 Under Learn, you will find Video walkthrough and PDF guide. 8 00:00:25,217 --> 00:00:28,289 We will practise first, then import a CSV file. 9 00:00:30,044 --> 00:00:31,559 Click Try sample data. 10 00:00:31,809 --> 00:00:35,969 The workspace opens with sixteen example rows and seven columns. 11 00:00:36,219 --> 00:00:39,333 Five numeric variables are selected for screening. 12 00:00:39,583 --> 00:00:45,237 The sample contains missing values and an unusual score so that you can explore the outputs. 13 00:00:45,487 --> 00:00:51,119 The left panel controls the analysis, and the tabs across the workspace show different results. 14 00:00:51,369 --> 00:00:53,822 You can import your own CSV at any time. 15 00:00:55,578 --> 00:01:02,127 For your own data, prepare a CSV with one header row, one case per row, and one variable per column. 16 00:01:02,377 --> 00:01:06,004 Click Import CSV in the Variables panel or in the menu. 17 00:01:06,254 --> 00:01:08,003 Choose the file from your device. 18 00:01:08,253 --> 00:01:15,784 In Select required variables, choose the numeric measures you intend to screen, then click Analyze selected variables. 19 00:01:16,034 --> 00:01:24,141 Do not treat participant identifiers or numeric category codes as continuous measures just because they contain numbers. 20 00:01:25,892 --> 00:01:28,580 After import, read the Screening Setting Notice. 21 00:01:28,830 --> 00:01:32,755 Then find the orange Data Screening Settings section in the left panel. 22 00:01:33,005 --> 00:01:36,675 Icons beside the titles help you recognise each setting. 23 00:01:36,925 --> 00:01:42,151 Missing-value handling decides whether incomplete cases are removed before screening. 24 00:01:42,401 --> 00:01:45,772 Z-scores and IQR screen one variable at a time. 25 00:01:46,022 --> 00:01:51,867 Mahalanobis distance screens unusual combinations of selected numeric values. 26 00:01:52,117 --> 00:01:55,488 Review Delete flagged rows from screened exports separately. 27 00:01:55,738 --> 00:01:59,471 Open a guide beside a setting when you need help choosing its cut-off. 28 00:01:59,721 --> 00:02:02,964 A flag is a prompt for review, not proof of an error. 29 00:02:04,725 --> 00:02:05,920 Open Overview. 30 00:02:06,170 --> 00:02:09,690 Rows is the number of cases in the current working dataset. 31 00:02:09,940 --> 00:02:13,246 Complete rows have values for every selected measure. 32 00:02:13,496 --> 00:02:16,718 Retained counts describe each workflow's screened export. 33 00:02:16,968 --> 00:02:21,768 Tested counts show the complete cases that received a Mahalanobis distance. 34 00:02:22,018 --> 00:02:25,474 These counts can differ when incomplete cases are retained. 35 00:02:25,724 --> 00:02:29,094 Expand Understanding overview counts for definitions. 36 00:02:29,344 --> 00:02:37,558 If a relevant detector is switched off, its cards indicate that the analysis was not applied rather than implying successful screening. 37 00:02:39,321 --> 00:02:41,006 Click Univar. Outliers. 38 00:02:41,256 --> 00:02:47,742 This page screens each selected variable separately using the enabled z-score and IQR rules. 39 00:02:47,992 --> 00:02:53,773 Review the summary table, the screening note, and the deleted and screened cases farther down the page. 40 00:02:54,023 --> 00:02:58,290 Check flagged values against the original records and your study context. 41 00:02:58,540 --> 00:03:01,889 If you change a rule in the left panel, the results update. 42 00:03:02,139 --> 00:03:06,427 Do not remove a valid observation solely because it is unusual. 43 00:03:08,184 --> 00:03:10,083 Click Multivar. Outliers. 44 00:03:10,333 --> 00:03:14,045 Review Mahalanobis Distance and its tested-case information. 45 00:03:14,295 --> 00:03:21,399 It looks for unusual combinations across the selected numeric variables, rather than a single extreme value. 46 00:03:21,649 --> 00:03:24,465 Scroll to Correlation Matrix on this same page. 47 00:03:24,715 --> 00:03:32,651 High-correlation cautions identify pairs worth investigating; the correlation matrix does not itself delete cases. 48 00:03:32,901 --> 00:03:41,263 Incomplete cases may have no Mahalanobis value, so distinguish tested cases from retained cases when reporting the results. 49 00:03:43,014 --> 00:03:45,617 Combined Outliers is a separate workflow. 50 00:03:45,867 --> 00:03:51,264 It applies univariate screening first, then Mahalanobis screening to the remaining cases. 51 00:03:51,514 --> 00:03:56,677 Its counts and data basis can therefore differ from standalone multivariate screening. 52 00:03:56,927 --> 00:04:00,809 Open Combined Outliers and review its summary and screening note. 53 00:04:01,059 --> 00:04:06,883 Use the combined export only when this sequential workflow is the one you intend to report. 54 00:04:07,133 --> 00:04:10,995 Do not assume that all screened exports contain the same cases. 55 00:04:12,750 --> 00:04:16,057 Click Normality and wait for the calculation to finish. 56 00:04:16,307 --> 00:04:21,896 First check Normality Assessment Data Basis so you know which cases each assessment uses. 57 00:04:22,146 --> 00:04:25,303 Expand Decision rule to read the app's criteria. 58 00:04:25,553 --> 00:04:32,444 Scroll to the univariate results, histograms and Q-Q plots, and the multivariate and combined assessments. 59 00:04:32,694 --> 00:04:37,814 Skewness and kurtosis describe distribution shape; they are not outlier- deletion rules. 60 00:04:38,064 --> 00:04:40,453 Read plots and supporting tests together. 61 00:04:40,703 --> 00:04:46,549 A normality label or one p-value does not settle every assumption of your planned analysis. 62 00:04:48,302 --> 00:04:51,779 Recommendations summarises the app's screening evidence. 63 00:04:52,029 --> 00:04:59,091 Start with Consolidated Decisions, then expand a relevant card under Recommended Statistical Analysis Path. 64 00:04:59,341 --> 00:05:03,799 The cards open independently, so you can focus on one topic at a time. 65 00:05:04,049 --> 00:05:07,527 Treat these recommendations as educational guidance. 66 00:05:07,777 --> 00:05:16,054 Your research question, measurement scales, study design, sample size and model assumptions still determine the final analysis. 67 00:05:16,304 --> 00:05:21,232 You can copy the reporting note as a draft, but review it before using it in a report. 68 00:05:22,991 --> 00:05:24,634 Imputation is optional. 69 00:05:24,884 --> 00:05:28,745 Open Imputation, choose a method, and click Run imputation. 70 00:05:28,995 --> 00:05:33,646 Here we demonstrate Mean only; it is not a recommendation for every study. 71 00:05:33,896 --> 00:05:40,594 Read the Imputation Report, then use Export imputed CSV to save the separate derived dataset. 72 00:05:40,844 --> 00:05:45,474 Use imputed data for screening switches the working dataset to that copy. 73 00:05:45,724 --> 00:05:48,134 Restore original data switches back. 74 00:05:48,384 --> 00:05:57,920 For important or substantial missingness, seek a justified missing-data strategy rather than treating simple replacement as a universal solution. 75 00:05:59,674 --> 00:06:02,213 Before leaving, save the outputs you need. 76 00:06:02,463 --> 00:06:10,847 Use Export screened CSV on the Univar., Multivar., or Combined Outliers page that matches your chosen workflow. 77 00:06:11,097 --> 00:06:22,339 Export cases CSV saves the case-review record; Export matrix CSV saves correlations; and Export normality CSV saves the normality outputs. 78 00:06:22,589 --> 00:06:30,653 Keep the original CSV, your screened or imputed copies, and a note of the selected variables, cut-offs and exclusions. 79 00:06:30,903 --> 00:06:36,642 Refreshing or closing the app clears its working data, so download your files before you finish.