ANDA'S IT LIBRARY
ABOUT ME

APPLIED PSYCHOLOGICAL SCIENCE LIBRARY
Here are some cross-referencing strategies among data science, psychological science and statistics, that I am exploring, as I feel they can be useful in real life:
Creating Apps
I have cross-referenced my pychological science library with the object-oriented programming structure used in online application development. Specifically, I have categorized behavioral phenomenon into programming clasess of attributes, methods and event triggers. This also easily conforms with prompt engineering, tapping into the vastness of AI. Overall, this opens the doors to so many new application possibilities in human behavior analysis and how we interact with our applications.
Big Data + Psychological Science
Cross-referencing company datasets of users with psychological science discoveries of behavior patterns; such as the psychometrics of trait analysis, and correlations of individual and group behaviours.
XAI + Cognitive Science
Cognitive science provides clues to current AI methods. On one hand, connectionist theories resemble neural nets, which are processed in parallel, making them fast, accurate, but not explainable. On the other hand, info-processing techniques are processed serially, are slower and sometimes less accurate, but we can clearly understand how the algorithm is processing individual cases. This is required for ethical and regulatory purposes and provides a foundation for Explainable Artificial Intelligence (XAI) applied to human behaviour.

I use R programming language for all my analysis. It's ecosystem is compatible with all other languages and data tools used in data science, statistics and psychological-social sciences. Therefore, it does not matter what my collegues are using, we all can still understand each other. For data users who do not have a background in data analysis, here is an explainer to get you started.

BEHAVIOURAL STATISTICS AND DATA MINING (2023)
The following is the scope of types of analysis tools I have scripted in R - organized in a way that works best for my brain:
WRANGLING
  1. Install and Load Libraries (install R | install packages | load packages)
  2. Import Data (csv | txt | xml | jso | rds | ovs (jamovi))
  3. Wrangle
    • a) filter cases & subgroups
    • b) recoding data
    • c) create scale scores
MODELING
Visualizing Data:
  • One Variable or Measure
    • creating - bar charts - shows counts, proportions
    • creating - histograms - shows distribution
    • creating - box plots - shows distribution
  • Multiple Variables and/or Measures
    • creating - scatterplots - shows relationship between 2 distributions
    • creating - multiple graphs - shows relationships amongst multiple distributions
    • creating - cluster charts - shows groupings of observations/cases
Exploring Data:
  • computing - frequencies
  • computing - descriptive statistics
  • computing - correlations
  • creating - contingency tables
Analysing Data:
  • comparing - proportions: chi-square for contingency tables
  • comparing - one mean to a population: one-sample t-test
  • comparing - paired means: paired-samples t-test
  • comparing - two means: independent samples t-test
  • comparing - multiple means: one-factor analysis of variance
  • comparing - means with multiple categorical predictors: factorial analysis of variance
Predicting Outcomes:
  • predicting - outcomes - with linear regression
  • predicting - outcomes - with lasso regression
  • predicting - outcomes - with Quantile regression
  • predicting - outcomes - with logistic regression
  • predicting - outcomes - with poisson or Log-linear regression
  • assessing - predictions - with blocked-entry models
DATA MINING (finding pattern in the noise)
ABOUT VARIABLES: Exploring Dimension Data
  • conducting - a principal component analysis
  • conducting - an item analysis
  • conducting - a confirmatory factor analysis
ABOUT CASES: Clustering and Classifying Observations/Cases:
  • grouping - cases - with hierarchical clustering
  • grouping - cases - with k-means clustering
  • classifying - cases - with k-nearest neighbors
  • classifying - cases - with decision tree analysis
  • creating - ensemble models - with random forest classification
ABOUT OTHER TYPES OF DATA STRUCTURES
  • predicting - classification behaviour - with association analysis
  • decomposing - dimension behavior over time - with time series analysis
  • grouping - word patterns - with text mining

gihub
back home