{
"cells": [
{
"cell_type": "markdown",
"id": "7bbbdeaa-39ca-49af-ba34-87a2d5e07813",
"metadata": {},
"source": [
"## 8 - Exploratory Factor Analysis\n",
"### Finding latent variables in data"
]
},
{
"cell_type": "markdown",
"id": "ee890cda-3b0f-4867-be51-97d7b8dc48e6",
"metadata": {},
"source": [
"We now start to move from the traditional modelling framework of using some variables to predict another, and into the realm of exploratory factor analysis (EFA). EFA is a *multivariate* technique which means that it has no clear distinction on what a 'predictor' or an 'outcome' variable is - rather, the aim of EFA is to take a set of variables and find another set of variables that represent the original ones. These are the so-called **latent** variables that, under the EFA model, are assumed to \"give rise\" to the original set of variables.\n",
"\n",
"EFA is commonly used to distil psychological constructs from things like questionnaire data. A popular example in psychology of this is the Big Five personality traits, which emerge from EFA of people rating adjectives about personalities to the extent they feel that are like that - e.g. \"outgoing\", \"anxious\".\n",
"\n",
"The key thing to remember is that EFA is also a linear model. You can think about it from the following somewhat quirky perspective - what an EFA does is takes each of your original variables, and tries to come up with a set of predictor variables that predicts each original variable well. It has to do that simultaneously for each of the original variables - so its really just coming up with a set of variables that 'best explain' all the observed variables at once!\n",
"\n",
"Lets see how we can fit one in Python. To do so, we'll use a new package, the `factor_analyzer` package, which has a model called `FactorAnalyzer`. Below we import a few key things as well as the new module."
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "a05336ae-3d46-4a5f-9e78-b41a4b7782aa",
"metadata": {},
"outputs": [],
"source": [
"# Import what we need\n",
"import pandas as pd # dataframes\n",
"import seaborn as sns # plots\n",
"import statsmodels.formula.api as smf # Models\n",
"import marginaleffects as me # marginal effects\n",
"import numpy as np # numpy for some functions\n",
"import pingouin as pg\n",
"from factor_analyzer import FactorAnalyzer # Note we write from factor_analyzer\n"
]
},
{
"cell_type": "markdown",
"id": "247795a9-71d3-4318-a32a-7ce2fd453625",
"metadata": {},
"source": [
"To examine this idea, we'll rely on a classic idea - that of the *g* factor of intelligence. The dataset below is taken from JASP's examples and comes from Spearman (1908) himself. It represents a set of scores on the following variables for a group of students:\n",
"\n",
"- Pitch - Score in pitch discrimination test.\n",
"- Light - Score in light discrimination test.\n",
"- Weight - Score in weight discrimination test.\n",
"- Classics - School grade for classic studies.\n",
"- French - School grade for French.\n",
"- English - School grade for English.\n",
"- Mathematics - School grade for mathematics.\n",
"\n",
"We can read in the data from the following link, then look at the top 5 rows. https://raw.githubusercontent.com/alexjonesphd/py4psy2024/refs/heads/main/G%20Factor.csv"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "3c993988-33ac-43ed-aaf7-1d4061ecfe7f",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Pitch
\n",
"
Light
\n",
"
Weight
\n",
"
Classics
\n",
"
French
\n",
"
English
\n",
"
Mathematics
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
50
\n",
"
10
\n",
"
4
\n",
"
16
\n",
"
19
\n",
"
10
\n",
"
7
\n",
"
\n",
"
\n",
"
1
\n",
"
3
\n",
"
10
\n",
"
6
\n",
"
5
\n",
"
6
\n",
"
6
\n",
"
5
\n",
"
\n",
"
\n",
"
2
\n",
"
10
\n",
"
10
\n",
"
6
\n",
"
5
\n",
"
6
\n",
"
6
\n",
"
5
\n",
"
\n",
"
\n",
"
3
\n",
"
60
\n",
"
10
\n",
"
9
\n",
"
22
\n",
"
23
\n",
"
22
\n",
"
22
\n",
"
\n",
"
\n",
"
4
\n",
"
4
\n",
"
12
\n",
"
5
\n",
"
1
\n",
"
1
\n",
"
1
\n",
"
2
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Pitch Light Weight Classics French English Mathematics\n",
"0 50 10 4 16 19 10 7\n",
"1 3 10 6 5 6 6 5\n",
"2 10 10 6 5 6 6 5\n",
"3 60 10 9 22 23 22 22\n",
"4 4 12 5 1 1 1 2"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Read in data\n",
"g = pd.read_csv('https://raw.githubusercontent.com/alexjonesphd/py4psy2024/refs/heads/main/G%20Factor.csv')\n",
"g.head(5)"
]
},
{
"cell_type": "markdown",
"id": "287c36a4-3cc7-4fd5-a9a8-0004d73ba87d",
"metadata": {},
"source": [
"Our goal is to find the underlying *g* variable - that is, 'intelligence' - a latent variable that gives rise to all of the scores we see across the tests. We do this by making a `FactorAnalyzer` model and fitting it to the data. Notice a key thing here - we deliberately set the number of factors to 1. This is a crucial consideration. In EFA you *must* specify the number of factors you want to see *before you start!*"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "d6b1d62c-2006-42a1-b3b6-cbc81c139589",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/opt/miniconda3/envs/py11/lib/python3.11/site-packages/factor_analyzer/factor_analyzer.py:663: UserWarning: No rotation will be performed when the number of factors equals 1.\n"
]
}
],
"source": [
"# Build and fit an EFA with a single factor\n",
"efa = FactorAnalyzer(n_factors=1).fit(g)"
]
},
{
"cell_type": "markdown",
"id": "ab2fa319-67eb-4da1-aa15-f72170d03949",
"metadata": {},
"source": [
"With this single model, there are many things to inspect to understand the overall result. Let's take a look at them!"
]
},
{
"cell_type": "markdown",
"id": "74f64afd-e056-47d5-80bf-6bd665369f30",
"metadata": {},
"source": [
"### Factor Loadings\n",
"The first key bit of information is how the original variables correlate with the discovered factor(s). In our example with *g*, we will see how each individual test correlates with the general measure of intelligence. These are often referred to as 'factor loadings', but don't let the term confuse you - they are correlations of the original variables with the discovered latent variables. We can obtain them from the `.loadings` attribute; here put into a dataframe to look nicer."
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "8d897996-1fff-426c-af43-5ef5091844f0",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
g_factor
\n",
"
\n",
" \n",
" \n",
"
\n",
"
Pitch
\n",
"
0.220054
\n",
"
\n",
"
\n",
"
Light
\n",
"
0.162702
\n",
"
\n",
"
\n",
"
Weight
\n",
"
0.094585
\n",
"
\n",
"
\n",
"
Classics
\n",
"
0.975240
\n",
"
\n",
"
\n",
"
French
\n",
"
0.939743
\n",
"
\n",
"
\n",
"
English
\n",
"
0.833758
\n",
"
\n",
"
\n",
"
Mathematics
\n",
"
0.911261
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" g_factor\n",
"Pitch 0.220054\n",
"Light 0.162702\n",
"Weight 0.094585\n",
"Classics 0.975240\n",
"French 0.939743\n",
"English 0.833758\n",
"Mathematics 0.911261"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Obtain loadings\n",
"g_loadings = pd.DataFrame(efa.loadings_,\n",
" columns=['g_factor'],\n",
" index=g.columns)\n",
"\n",
"g_loadings"
]
},
{
"cell_type": "markdown",
"id": "12db62fd-1f1a-4b75-a652-f41256e6145f",
"metadata": {},
"source": [
"This illustrates scores on mathematics, classics, and French tests are all *very* strongly correlated with the discovered factor. Conversely, the weight and light tests are poorly correlated. That is, the general intelligence variable does not influence them much.\n",
"\n",
"**Convention suggests that if a variable has a correlation with a factor of greater than .4 or less than -.4, it is \"loaded\" on that factor**."
]
},
{
"cell_type": "markdown",
"id": "b5a9257e-280f-4ab7-bc02-c9a956eafb50",
"metadata": {},
"source": [
"### Factor variances\n",
"Another important bit of information EFA gives us is the *explained variance* of the factors. The latent variables will, depending on the data and number of them, do a more or less good job of explaining th variance in the *original variables*. \n",
"\n",
"We can obtain these from our models `.get_factor_variance()` method, which returns a list of three types - variance of factors (less useful), variance due to each factor (very handy) and cumulative variance. In our case here there is only one value, but later examples will see more factors and thus more variance."
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "20012b8d-55a4-4309-bd05-78bef92d02a3",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(array([3.44360164]), array([0.49194309]), array([0.49194309]))"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Get factor variances\n",
"efa.get_factor_variance() \n",
"# 1. Variance of each factor\n",
"# 2. Amount of variance each factor is responsible for in the data\n",
"# 3. The cumulative variance across the factors"
]
},
{
"cell_type": "markdown",
"id": "be05feca-d8e3-4a2f-9d6c-d886573a48fe",
"metadata": {},
"source": [
"### Communalities\n",
"This is another example of statisticians using weird words to describe a concept we already know about. Each original variable has a 'communality' value - this is simply the amount of variance the factors explain in that variable - remember, an EFA is 'figuring out' the *predictors* of a regression for each of the variables. A good EFA model has high communalities for all variables, typically above .7. In using EFA to build psychometric tests we might drop questions with low communalities. We get communalities like so:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "9ff8141f-4b71-4389-b5d8-d762a7c74eac",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
communality
\n",
"
\n",
" \n",
" \n",
"
\n",
"
Pitch
\n",
"
0.048424
\n",
"
\n",
"
\n",
"
Light
\n",
"
0.026472
\n",
"
\n",
"
\n",
"
Weight
\n",
"
0.008946
\n",
"
\n",
"
\n",
"
Classics
\n",
"
0.951093
\n",
"
\n",
"
\n",
"
French
\n",
"
0.883117
\n",
"
\n",
"
\n",
"
English
\n",
"
0.695152
\n",
"
\n",
"
\n",
"
Mathematics
\n",
"
0.830397
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" communality\n",
"Pitch 0.048424\n",
"Light 0.026472\n",
"Weight 0.008946\n",
"Classics 0.951093\n",
"French 0.883117\n",
"English 0.695152\n",
"Mathematics 0.830397"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Get communalities\n",
"comms = pd.DataFrame(efa.get_communalities(), columns=['communality'], index=g.columns)\n",
"comms"
]
},
{
"cell_type": "markdown",
"id": "f5342382-ce3a-4ff1-a8f0-fbfda913fca8",
"metadata": {},
"source": [
"These look OK for classics, french and mathematics, but poorer for the others - in this simple case the interpretation is similar to the loadings we saw earlier."
]
},
{
"cell_type": "markdown",
"id": "559bcb35-e598-49d1-85e1-6b27aed3c482",
"metadata": {},
"source": [
"### The latent factors themselves\n",
"Finally, if we want to, we can obtain an actual estimate of the latent variable itself from the model. This is useful to examine or to even perhaps include in another model. We do this by asking our EFA model to `transform` the original data, whereupon we get a representation of the latent variable(s). In this case, we obtan scores on *g*, which look like a Z-scored variable, with higher numbers indicating more intelligence."
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "32c585ca-59c1-4a18-a9b1-a36441c40079",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
g_factor
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
0.255022
\n",
"
\n",
"
\n",
"
1
\n",
"
-1.004152
\n",
"
\n",
"
\n",
"
2
\n",
"
-1.011543
\n",
"
\n",
"
\n",
"
3
\n",
"
1.510647
\n",
"
\n",
"
\n",
"
4
\n",
"
-1.594142
\n",
"
\n",
"
\n",
"
5
\n",
"
-1.303185
\n",
"
\n",
"
\n",
"
6
\n",
"
0.282946
\n",
"
\n",
"
\n",
"
7
\n",
"
1.690320
\n",
"
\n",
"
\n",
"
8
\n",
"
-0.255428
\n",
"
\n",
"
\n",
"
9
\n",
"
-1.250707
\n",
"
\n",
"
\n",
"
10
\n",
"
-0.752566
\n",
"
\n",
"
\n",
"
11
\n",
"
1.069516
\n",
"
\n",
"
\n",
"
12
\n",
"
-0.096286
\n",
"
\n",
"
\n",
"
13
\n",
"
-1.396858
\n",
"
\n",
"
\n",
"
14
\n",
"
-0.135562
\n",
"
\n",
"
\n",
"
15
\n",
"
0.692286
\n",
"
\n",
"
\n",
"
16
\n",
"
1.349655
\n",
"
\n",
"
\n",
"
17
\n",
"
1.112337
\n",
"
\n",
"
\n",
"
18
\n",
"
0.828171
\n",
"
\n",
"
\n",
"
19
\n",
"
0.383408
\n",
"
\n",
"
\n",
"
20
\n",
"
-0.237878
\n",
"
\n",
"
\n",
"
21
\n",
"
0.668323
\n",
"
\n",
"
\n",
"
22
\n",
"
-0.804324
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" g_factor\n",
"0 0.255022\n",
"1 -1.004152\n",
"2 -1.011543\n",
"3 1.510647\n",
"4 -1.594142\n",
"5 -1.303185\n",
"6 0.282946\n",
"7 1.690320\n",
"8 -0.255428\n",
"9 -1.250707\n",
"10 -0.752566\n",
"11 1.069516\n",
"12 -0.096286\n",
"13 -1.396858\n",
"14 -0.135562\n",
"15 0.692286\n",
"16 1.349655\n",
"17 1.112337\n",
"18 0.828171\n",
"19 0.383408\n",
"20 -0.237878\n",
"21 0.668323\n",
"22 -0.804324"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Obtain latent scores\n",
"latent = efa.transform(g)\n",
"\n",
"# Put it in a dataframe\n",
"latent = pd.DataFrame(latent, columns=['g_factor'])\n",
"latent"
]
},
{
"cell_type": "markdown",
"id": "fc6ea7e0-1582-49d1-bb92-eda0197bae34",
"metadata": {},
"source": [
"### Psychometrics and EFA\n",
"A common use case for EFA is in the developmental of psychometric tests or questionnaires. Researchers can use EFA uncover a latent set of dimensions underpinning responses to a questionnaire. In this way they can refine the test by dropping some items, including more, or establishing how many 'things' their test is measuring (i.e. is it measuring 5 aspects of personality, or a single construct like intelligence?).\n",
"\n",
"Lets see an example of a larger dataset with more than one latent factor - specifically, the Big Five personality variables. This dataset contains responses from 2,800 people on the Big Five Inventory, a questionnaire measuring personality. There are 25 questions, each asking the person to say how much they agree with a particular statement. The EFA model thus assumes there are several latent variables that give rise to peoples responses on these questions, and our aim is to find them.\n",
"\n",
"The specific questions are available [here](https://vincentarelbundock.github.io/Rdatasets/doc/psych/bfi.html), along with the data at this link: https://vincentarelbundock.github.io/Rdatasets/csv/psych/bfi.csv\n",
"\n",
"First, we can read it in and display it. If you take a closer look at the columns, you can see there are five questions for each trait, with each question denoted by an initial and a number, e.g. E1 is the first question measuring Extraversion, O2 the second for Openness, and so on."
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "5c900dcc-a250-44cd-a4a0-4aaf472d957c",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Visualise heatmap\n",
"sns.heatmap(big5_loadings, \n",
" cmap='Greys', # Greyscale colour, so black-to-white\n",
" annot=True, # Labels with the actual correlation\n",
" fmt='.2f') # Tells the label to be 2 decimal places"
]
},
{
"cell_type": "markdown",
"id": "95530bc5-dbfe-4778-baa0-a0dde3ccc83e",
"metadata": {},
"source": [
"This is an interesting result. We can see now, clearly, the existence of the factors. For example, `fac1` has high loadings with the five \"N\" questions, indicating it to be Neuroticism. Similarly, `fac2` has high loadings on the \"E\" questions, suggesting Extraversion, and so on. Notice some of the other correlations within the factors are relatively high with other types of questions too - its never a clear cut solution.\n",
"\n",
"We can examine our factor variances:"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "d78084fd-c04a-4ffb-ae4e-dfb5517e6e6d",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(array([2.70387986, 2.48623235, 2.04986716, 1.63844108, 1.46150922]),\n",
" array([0.10815519, 0.09944929, 0.08199469, 0.06553764, 0.05846037]),\n",
" array([0.10815519, 0.20760449, 0.28959917, 0.35513682, 0.41359719]))"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Show factor variances\n",
"big5_efa.get_factor_variance()"
]
},
{
"cell_type": "markdown",
"id": "8353acea-89e1-4dee-936c-eabec8e00e2e",
"metadata": {},
"source": [
"So taken together, all factors capture about half the variance in the data. The first factor (which we think is Neuroticism...!) explains about 10%, the second, Extraversion, explains about 9%, and so on. \n",
"\n",
"What about the communalities? Do any of these indicate anything noteworthy? Again here it is helpful to visualise these."
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "f69a1e11-cb76-4575-9d45-2296c3dc6bb9",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Visualise communalities\n",
"comms = pd.DataFrame(big5_efa.get_communalities(), index=bf3.columns, columns=['Communality'])\n",
"\n",
"# Show\n",
"sns.heatmap(comms, cmap='Greys', annot=True)"
]
},
{
"cell_type": "markdown",
"id": "36ca6f52-4ad0-49e7-b428-34a9bae81390",
"metadata": {},
"source": [
"Thse don't look *too* bad. There are no hard and fast rules about what's good here, other than higher is better. Some of these like A4 and O4 may raise some eyebrows, but in general these would be acceptable. EFA is more an art than a science!\n"
]
},
{
"cell_type": "markdown",
"id": "fc139ee9-fcfa-4f27-a898-ecd289c81b32",
"metadata": {},
"source": [
"Its now possible to go one step further and examine the correlations amongst the factors themselves! That is, are the personality traits themselves correlated? This is only possible because we let `FactorAnalyzer` allow them to be correlated - there are many factor analysis approaches that produce completely uncorrelated factors, but as a rule, correlated factors make sense. These are stored in the `.phi_` attribute of the model."
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "e1114e29-e51e-4bfd-a1f9-c6bb574019dc",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
neuro
\n",
"
extra
\n",
"
consci
\n",
"
agree
\n",
"
open
\n",
"
\n",
" \n",
" \n",
"
\n",
"
neuro
\n",
"
1.000000
\n",
"
-0.256314
\n",
"
0.398860
\n",
"
0.142306
\n",
"
0.345106
\n",
"
\n",
"
\n",
"
extra
\n",
"
-0.256314
\n",
"
1.000000
\n",
"
-0.224364
\n",
"
0.040343
\n",
"
-0.013000
\n",
"
\n",
"
\n",
"
consci
\n",
"
0.398860
\n",
"
-0.224364
\n",
"
1.000000
\n",
"
0.191999
\n",
"
0.236255
\n",
"
\n",
"
\n",
"
agree
\n",
"
0.142306
\n",
"
0.040343
\n",
"
0.191999
\n",
"
1.000000
\n",
"
0.155367
\n",
"
\n",
"
\n",
"
open
\n",
"
0.345106
\n",
"
-0.013000
\n",
"
0.236255
\n",
"
0.155367
\n",
"
1.000000
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" neuro extra consci agree open\n",
"neuro 1.000000 -0.256314 0.398860 0.142306 0.345106\n",
"extra -0.256314 1.000000 -0.224364 0.040343 -0.013000\n",
"consci 0.398860 -0.224364 1.000000 0.191999 0.236255\n",
"agree 0.142306 0.040343 0.191999 1.000000 0.155367\n",
"open 0.345106 -0.013000 0.236255 0.155367 1.000000"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Get correlations of all factors\n",
"factor_corrs = big5_efa.phi_\n",
"\n",
"# Tidy it up into a DataFrame, with the suspected names of the factors\n",
"factor_corrs = pd.DataFrame(factor_corrs, \n",
" columns=['neuro', 'extra', 'consci', 'agree', 'open'],\n",
" index=['neuro', 'extra', 'consci', 'agree', 'open'])\n",
"\n",
"factor_corrs"
]
},
{
"cell_type": "markdown",
"id": "230393d6-6675-41e3-9122-616fb1892c99",
"metadata": {},
"source": [
"Neurotic people are less extraverted but more conscientious, for example. Finally, we can give each person a score on the factors too, by extracting the latent variable score for each person. This would be useful to actually score someones Big 5 traits in terms of the latent variable."
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "308b26d5-ece5-41fc-8ab0-83d6b492b038",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
neuro
\n",
"
extra
\n",
"
consci
\n",
"
agree
\n",
"
open
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
-0.368754
\n",
"
-0.298491
\n",
"
-1.324973
\n",
"
-0.917815
\n",
"
-1.661550
\n",
"
\n",
"
\n",
"
1
\n",
"
0.081186
\n",
"
0.352849
\n",
"
-0.535773
\n",
"
-0.176275
\n",
"
-0.278700
\n",
"
\n",
"
\n",
"
2
\n",
"
0.496539
\n",
"
0.106411
\n",
"
-0.047870
\n",
"
-0.658264
\n",
"
0.194505
\n",
"
\n",
"
\n",
"
3
\n",
"
-0.136226
\n",
"
-0.169679
\n",
"
-1.040085
\n",
"
-0.115630
\n",
"
-1.110601
\n",
"
\n",
"
\n",
"
4
\n",
"
-0.369298
\n",
"
0.230058
\n",
"
-0.101516
\n",
"
-0.810210
\n",
"
-0.725543
\n",
"
\n",
"
\n",
"
5
\n",
"
0.016293
\n",
"
1.238686
\n",
"
1.446943
\n",
"
0.204831
\n",
"
0.469544
\n",
"
\n",
"
\n",
"
6
\n",
"
-1.188218
\n",
"
0.204278
\n",
"
0.182627
\n",
"
0.137170
\n",
"
0.776679
\n",
"
\n",
"
\n",
"
7
\n",
"
0.547863
\n",
"
-1.806800
\n",
"
-1.227558
\n",
"
-2.100190
\n",
"
-0.616149
\n",
"
\n",
"
\n",
"
8
\n",
"
0.742044
\n",
"
0.847145
\n",
"
1.193344
\n",
"
0.391977
\n",
"
0.263996
\n",
"
\n",
"
\n",
"
9
\n",
"
-0.188935
\n",
"
0.247622
\n",
"
-0.144489
\n",
"
-0.078716
\n",
"
-0.329985
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" neuro extra consci agree open\n",
"0 -0.368754 -0.298491 -1.324973 -0.917815 -1.661550\n",
"1 0.081186 0.352849 -0.535773 -0.176275 -0.278700\n",
"2 0.496539 0.106411 -0.047870 -0.658264 0.194505\n",
"3 -0.136226 -0.169679 -1.040085 -0.115630 -1.110601\n",
"4 -0.369298 0.230058 -0.101516 -0.810210 -0.725543\n",
"5 0.016293 1.238686 1.446943 0.204831 0.469544\n",
"6 -1.188218 0.204278 0.182627 0.137170 0.776679\n",
"7 0.547863 -1.806800 -1.227558 -2.100190 -0.616149\n",
"8 0.742044 0.847145 1.193344 0.391977 0.263996\n",
"9 -0.188935 0.247622 -0.144489 -0.078716 -0.329985"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Extract the scores\n",
"traits = big5_efa.transform(bf3)\n",
"\n",
"# Into a dataframe\n",
"traits = pd.DataFrame(traits,\n",
" columns=['neuro', 'extra', 'consci', 'agree', 'open'])\n",
"\n",
"# show traits\n",
"traits.head(10)"
]
},
{
"cell_type": "markdown",
"id": "983afd9b-c8f3-4f08-bf93-4bcacf4685cc",
"metadata": {},
"source": [
"As an example, participant 5 has some particularly high scores on extrversion and conscientiousness, while participant 0 has rather low openness."
]
},
{
"cell_type": "markdown",
"id": "91521a88-c5e5-4828-98de-136b7bcd5314",
"metadata": {},
"source": [
"### But **how** many factors are there?!\n",
"We come now to the crux of EFA, and a point that causes dramatic discussion in the academic literature. For example you may have heard of the HEXACO model of personality (six factors) as opposed to the Big 5 - the same kinds of data can produce different numbers of factors. How do we choose the number of factors to extract? In the above examples we had a good theoretical grounding that guided us, but often times we are using EFA in an *exploratory* sense, that is, to explore the possible latent variables in the data.\n",
"\n",
"There are many, many techniques for determining the number of factors in a dataset, and all of them come with caveats and controversy. This is - and has been for decades - an active area of research in statistics. We will look at one technique here which does reasonably well at identifying the \"right\" number of factors, and has a sensible philosophical background, called *parallel analysis*.\n",
"\n",
"How it works is technical, but the basic idea goes like this:\n",
"\n",
"1. Looking at the data you have collected, create a set of randomly-generated data that is the same number of rows and columns that is basically entirely uncorrelated.\n",
"2. Conduct a factor analysis on this created data, with the number of factors set to the number of variables (e.g. if 10 variables, get 10 factors), and calculate the 'eigenvalues' - these represent how much variance each factor explains.\n",
"3. Repeat step 2 thousands of times and collect all the variances explained for each factor.\n",
"4. Now in your real data, conducted the factor analysis with the number of factors equal to the number of variables. Get the variances explained.\n",
"5. Find where the variances in your actual data are *above and beyond* those seen from step 3. Count along until the variances of the real data *are within the bounds* of those at step 3. The number of factors up to that point is a good guess of how many factors there are, simply because beyond that the values are consistent with *entirely made up nonsense data!*\n",
"\n",
"That sounds complicated, but its really using the idea of what you'd expect to see by chance as a way to understand what you see your data. Fortunately, conducting it in Python is very simple.\n",
"\n",
"We use the `parallel_analysis` function from the `horns` package. We specify the analysis type as 'fa' (this approach also works for other types of analyses in this domain, like principal components analysis, so we are explicit to tell it we want an EFA), ask for the full output, and we have to convert our DataFrame to a `numpy` array. We also limit it to 500 repeats, as more take longer!\n",
"\n",
"Its as simple as the below:"
]
},
{
"cell_type": "code",
"execution_count": 18,
"id": "88022110-e0f0-4a94-9c06-e4bd456e54b3",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n",
"OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Import\n",
"from horns import parallel_analysis\n",
"\n",
"# Conduct\n",
"pa_results = parallel_analysis(bf3.to_numpy(), simulations=500, analysis_type='fa', full_output=True)"
]
},
{
"cell_type": "markdown",
"id": "e1846ade-dda3-4de4-9f1a-d9a3d74395d1",
"metadata": {},
"source": [
"Parallel analysis returns a plot and a dictionary of results. The plot shows the cutoff points at which number of factors are most 'surprising' given a dataset of this size and shape. We can access the number of factors as follows:"
]
},
{
"cell_type": "code",
"execution_count": 19,
"id": "077e3977-29ff-4635-ad09-1c3442c5908e",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"8\n"
]
}
],
"source": [
"# Print number of factors\n",
"print(pa_results['factors'])"
]
},
{
"cell_type": "markdown",
"id": "7cadb327-14e2-40d1-a74d-35811db41868",
"metadata": {},
"source": [
"So despite us selecting out 5, parallel analysis suggests there *eight* latent factors in the data! \n",
"\n",
"EFA is as much an art as it is a science (much like other forms of modelling), so take care when conducting it. Blindly following one approach (like top-down theory) or another (bottom-up approaches like parallel analysis) can lead you astray. To finish, let's see what an 8-factor solution of personality looks like:"
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "36b0c163-a345-4752-9b33-b1fca2d986e9",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# 8 factors\n",
"big8_efa = FactorAnalyzer(n_factors=8).fit(bf3)\n",
"\n",
"# Examine the loadings\n",
"loadings_8 = pd.DataFrame(big8_efa.loadings_, index=bf3.columns, columns=['fa1', 'fa2', 'fa3', 'fa4', 'fa5', 'fa6', 'fa7', 'fa8'])\n",
"\n",
"# Heatmap\n",
"sns.heatmap(loadings_8, cmap='Greys', annot=True, fmt='.2f')"
]
},
{
"cell_type": "markdown",
"id": "a23160a1-c1f9-4e63-a5e4-b49becaeeff4",
"metadata": {},
"source": [
"Notice here that while traits like Neuroticism and Openness orrelate with one factor (number 1 and 4 respectively), there are 'dilutions' of other traits across factors - Extraversion is spread across factors 2 and 5, and some aspects of it onto factor 8. This would be a cause for concern for psychometricians hoping for a neat set of questions measuring 5 latent factors and might result in a re-thinking of certain questions. "
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.12.2"
}
},
"nbformat": 4,
"nbformat_minor": 5
}