Testing Conditional Independence in Psychometric Networks:

An Analysis of Three Bayesian Methods

Sekulovski, N.1 , Keetelaar, S.1 , Huth, K.B.S.1,2,3 , Wagenmakers, E-J.1 ,

van Bork, R.1 , van den Bergh, D.1,4 , & Marsman, M.1,3,4
1 Department of Psychology, University of Amsterdam;
2 Department of Psychiatry, Amsterdam UMC location, University of Amsterdam;
3 Centre for Urban Mental Health, University of Amsterdam;
4 Amsterdam Brain and Cognition, University of Amsterdam

The paper is now published at Multivariate Behavioral Research. Please cite

the published version:
https://www.tandfonline.com/doi/full/10.1080/00273171.2024.2345915.

Abstract
Network psychometrics uses graphical models to assess the network structure
of psychological variables. An important task in their analysis is determin-
ing which variables are unrelated in the network, i.e., are independent given
the rest of the network variables. This conditional independence structure
is a gateway to understanding the causal structure underlying psychological
processes. Thus, it is crucial to have an appropriate method for evaluating
conditional independence and dependence hypotheses. Bayesian approaches
to testing such hypotheses allow researchers to differentiate between absence
of evidence and evidence of absence of connections (edges) between pairs of
variables in a network. Three Bayesian approaches to assessing conditional
independence have been proposed in the network psychometrics literature.
We believe that their theoretical foundations are not widely known, and
therefore we provide a conceptual review of the proposed methods and high-
light their strengths and limitations through a simulation study. We also
illustrate the methods using an empirical example with data on Dark Triad
Personality. Finally, we provide recommendations on how to choose the opti-
mal method and discuss the current gaps in the literature on this important
topic.
Word count: 10023 words

Keywords: Bayesian model averaging, Bayes factor, Conditional indepen-

dence, Markov random fields
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 2

Introduction

In network psychometrics, graphical models known as Markov Random Fields (Kin-

dermann & Snell, 1980; Rozanov, 1982) have become a popular tool for assessing the network
structure of psychological variables (see Borsboom et al., 2021; Contreras, Nieto, Valiente,
Espinosa, & Vazquez, 2019; Marsman & Rhemtulla, 2022; Robinaugh, Hoekstra, Toner, &
Borsboom, 2020, for recent reviews). In these networks, nodes represent observed psycho-
logical variables (e.g., symptom indicators), and edges represent the pairwise relationships
between them. A ‘present’ relationship indicates a direct link (positive or negative) between
a pair of variables, while an absent relationship reflects conditional independence; if any
dependence exists it is fully accounted for by other variables in the network. For example, al-
though shark attacks (A) and ice cream sales (I) are correlated, their dependence disappears
when season (S) is taken into account, showing that A and I are conditionally independent
given S. The goal of network analysis is to discover the underlying configuration of present
and absent edges — the conditional independence structure of the network. Conditional
independence is a first step in determining causal relationships (e.g., Pearl, 2009; Spirtes,
Glymour, & Scheines, 2000), and aids our understanding of the underlying dynamic system.
With the gradual development of Bayesian approaches to network analysis (e.g., Marsman
& Haslbeck, 2023; Marsman, Maris, Bechger, & Glas, 2015; A. Mohammadi & Wit, 2015;
Williams, 2021; Williams & Mulder, 2020a), testing the conditional independence of a pair
of variables in the network has also begun to receive attention: Three Bayesian approaches
to conditional independence testing have recently been proposed in the network psychomet-
rics literature. Although the development of conditional independence tests is an important
step forward, the three methods differ conceptually, and we believe that their foundations
are not well known. In this paper, we provide a conceptual review of the three Bayesian
approaches, and show that when we are uncertain about the structure of the network, there
appears to be an optimal approach for testing conditional independence.
In recent years, both frequentist and Bayesian approaches have been proposed for es-
timating the network structure from empirical data. In psychology, frequentist approaches
are the norm, and although unregularized estimation approaches have become available in
recent years (see Blanken, Isvoranu, & Epskamp, 2022, for a review), regularized estima-
tion based on lasso remains popular (e.g., Borsboom et al., 2021; Epskamp & Fried, 2018;
Tibshirani, 1996; van Borkulo et al., 2014). A caveat of these approaches is that they are
biased toward the null hypothesis of conditional independence, yet their underlying frequen-
tist methodology can only refute this hypothesis, not support it. This leads to the problem
that if an edge is missing from an estimated network, we are unsure whether this is due to a
lack of information in the data to support the relation, or due to actual conditional indepen-
dence (e.g., Epskamp, Kruis, & Marsman, 2017). Bayesian methods (e.g., Wagenmakers,
Marsman, et al., 2018; Wagenmakers, Morey, & Lee, 2016), on the other hand, are able to
quantify the relative support for the competing hypotheses of conditional dependence and
conditional independence; importantly, they can also reveal the lack of support for either

Correspondence concerning this article should be addressed to Nikola Sekulovski, University of Am-
sterdam, Psychological Methods, Nieuwe Achtergracht 129B, PO Box 15906, 1001 NK Amsterdam, The
Netherlands. E-mail: [email protected].
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 3

hypothesis in the data at hand. By distinguishing evidence of absence from absence of evi-
dence, Bayesian methods facilitate a deeper understanding of the conditional independence
structure of the network.
Below we review three Bayesian approaches that have recently been proposed to test
conditional independence. The first method uses the credible interval – the Bayesian version
of the frequentist confidence interval – and assesses whether or not it contains the parameter
values that indicate conditional independence. This method focuses solely on rejecting the
conditional independence hypothesis, and thus suffers from the same fundamental problem
that plagues the frequentist methods mentioned above. The second method uses a Bayes
factor approach (Jeffreys, 1961; Kass & Raftery, 1995) —which is the Bayesian generaliza-
tion of the likelihood ratio test. The Bayes factor compares how well two competing models
can predict the observed data. When we compare two models that are identical except that
one has two variables that are unrelated and the other has them related, we can use the
Bayes factor to express the relative support, or lack thereof, for the conditional dependence
or independence hypotheses. The Bayes factor test represents a major improvement over
interval-based tests for conditional dependence and independence. However, we will show
that this Bayes factor approach requires a choice about which relationships are present in
the rest of the network and that it is sensitive to that choice. The third method, called the
inclusion Bayes factor, is a generalization of the Bayes factor approach that uses Bayesian
model averaging (BMA, Hoeting, Madigan, Raftery, & Volinsky, 1999; Kaplan, 2021) to
overcome the sensitivity to which relationships are present in the rest of the network. The
inclusion Bayes factor compares how well we can predict the observed data from a com-
bination of all models in which the two variables are related, and compares this to the
predictive adequacy of a combination of models in which the variables are unrelated. In
this paper, we consider the structure of the network, a particular configuration of present
and absent edges, to be a model. As we will show below, BMA allows us to make robust
structure-averaged inferences.
The remainder of this paper is structured as follows. The next two sections pro-
vide a conceptual introduction to the role of conditional dependence and independence in
graphical modeling, and the Bayesian methodology that underlies the methods for testing
these hypotheses. We refer the interested reader to Epskamp, Haslbeck, Isvoranu, and van
Borkulo (2022), Marsman et al. (2018), and Waldorp and Marsman (2022) for a detailed
introduction to the graphical models used in network psychometrics, to van de Schoot et al.
(2014) and Wagenmakers, Marsman, et al. (2018) for a detailed introduction to Bayesian
estimation and hypothesis testing, and to Huth, de Ron, et al. (2023) for a more compre-
hensive introduction to the Bayesian analysis of graphical models. In the third section, we
investigate the three Bayesian methods to test for conditional dependence and independence
in detail and discuss their relations and limitations, after which we compare their relative
performance in a simulation study. We end by discussing the limitations of the Bayesian
analysis of graphical models, and Bayesian model averaging in particular.

Graphical Modeling
A graphical model specifies the joint probability distribution for a set of observed
variables, and represents these variables as nodes in a network. The goal of a statistical
analysis of the graphical model is to determine the relations between pairs of variables,
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 4

which will constitute the edges of the network. We usually have two questions about
network relations. First, we wish to know if the edge is there or not: is an effect present?
Once we have established that there is an effect, and the edge should be in the model, a
follow-up question could be how strong the relation is: what is the strength of the effect?
The first question is usually linked to testing whereas the latter is linked to estimation, but
this distinction can be vague in practice.
In this paper, we will address the two questions about the graphical model separately.
First, a binary variable γij is used to indicate that the edge between variables i and j is
present (i.e., γij = 1) or absent (i.e., γij = 0). In a network with p variables, we have
k = p (p−1)/2 possible edges. Each configuration of edges (i.e., the pattern of zeros and
ones: γ12 , . . . , γ(p−1) p ) constitutes a possible network structure Ss , of which there are
2k in total. Figure 1 illustrates the idea for a network of three random variables; shark
attacks (A), ice cream sales (I), and season (S). The three variables yield k = 3×(3−1)/2 = 3
possible edges, and thus 2k = 23 = 8 possible network structures. For example, Structure
S1 = [γAI = 0, γAS = 0, γSI = 0] and Structure S4 = [γAI = 0, γAS = 0, γSI = 1].

Figure 1 . The possible structures along with their Posterior Structure Probabilities for the
random variables: shark attacks (A), ice cream sales (I), and season (S)

For each structure Ss , s = 1, . . . , 2k , we have a distinct statistical model1 for the

observed variables p(data | Θs , Ss ). The weights of the relations in the structure Ss are
expressed in a symmetric matrix Θs , which has the edge weights θij on the off-diagonals.
The edge weights that correspond to relations that are absent from Ss are set to zero
in Θs . For the graphical models that we analyze in this paper —Markov Random Field
(MRF) models (Kindermann & Snell, 1980; Rozanov, 1982)— these edge weights are partial
1
In the Bayesian literature, the model is usually indicated with Ms , but here we use Ss to connect it to
the network’s structure.
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 5

associations, which indicate the strength of the relation between two variables that excludes
the influence of other variables in the model. The higher the absolute value of the partial
association θij , the stronger the two variables influence each other.
Several MRF models are used in network psychometrics, which differ primarily in
the level of measurement of the variables. For example, the Ising model (Ising, 1925)
is a graphical model for binary variables (e.g., symptom indicators), the ordinal MRF
(Marsman & Haslbeck, 2023) extends the Ising model to also include ordinal variables, the
Gaussian graphical model (GGM; Lauritzen, 2004) is used for continuous variables, and
the mixed graphical model (MGM; Haslbeck & Waldorp, 2020) handles binary, unordered
categorical, count, and continuous variables. For the particular case of the GGM, the
matrix Θ is known as the precision matrix, and when standardized, it contains partial
correlations (e.g., Waldorp & Marsman, 2022). To keep the discussion general, we will
refer to the elements of Θ as edge weights. Other graphical models that are also used in
the network psychometrics literature but are not MRF models are the multivariate ordered
probit (Guo, Levina, Michailidis, & Zhu, 2015) for ordinal variables and the Gaussian
copula graphical model (Dobra & Lenkoski, 2011) for mixed binary, ordinal, and continuous
variables. However, in this paper we focus exclusively on MRF models.

Conditional Independence

In network psychometrics, it is often assumed that the observed data are variables in
a complex, dynamic system. The underlying system has a causal component in that some
variables influence other variables in a particular way, and some of these relationships are
reciprocal. Since it is difficult to learn the directed, causal relationships from correlational
data, we use undirected graphical models to model the relationships among the variables
in the underlying system. MRFs are an important class of undirected graphical models be-
cause their parameters tell us directly about the conditional dependence and independence
between variables in the network: If the edge weight θij between variables i and j is zero,
then the two variables are conditionally independent. MRFs are thus convenient models
for assessing conditional independence, and, since conditional independence is a gateway to
learning the underlying causal structure (e.g., Pearl, 2009; Spirtes et al., 2000), they play
an important role in the graphical approach to causal inference (Ryan, Bringmann, & Schu-
urman, 2022). One could, of course, adopt a purely statistical interpretation of conditional
independence without considering potential causal implications. However, since the notion
of conditional independence is also central to causal inference, we wish to clarify how the
two are related in this subsection.
Spirtes et al. (2000), Pearl (2009), and others (see Glymour, Zhang, & Spirtes, 2019,
for a recent overview) have developed the graphical approach to causal inference as a formal
framework in which causal relationships are represented as directed acyclic graphs (DAGs).2
Conditional dependencies and independencies are key to identifying DAGs that are consis-
tent with observed data. For example, consider the three variables A, S, and I in Figure
1. From their correlations alone, we cannot identify causal relationships among the three
variables. However, if we also knew their conditional dependencies, e.g., that A and I are
2
DAGs are sometimes referred to as Bayesian networks. We wish to emphasize that Bayesian networks
(DAGs) are different from Bayesian analysis of (MRF) graphical models, which is the focus of this paper.
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 6

conditionally independent given S, while A and S, and A and I are conditionally dependent
(i.e., A–S–I, such as S6 in Figure 1), we could take a step toward causal discovery. Under
some strong assumptions (e.g., there are no unobserved confounders, there is no selection
bias, and the causal relations do not cancel each other out; Eberhardt, 2017), one can use
the conditional independence structure S to infer three possible (directed) causal graphs:
A → S → I, A ← S ← I, and A ← S → I. For a detailed introduction to learning causal
relations from conditional dependence and independence, we refer the interested reader to
Pearl (2009).
The conditional independence structure S is a middle ground between simple un-
conditional associations and directed causal graphs: Simple associations will contain many
spurious relations that disappear when conditioning on other variables in the network.
While this conditioning removes associations that can be explained through other variables
in the network, it can also induce spurious relations: any variable that is a common effect
of other variables in the network will induce a spurious association between these variables
when conditioned on. It is therefore important to note that not all conditional dependen-
cies will reflect causal relations unless strong assumptions are made (such as the absence of
common effects and unobserved common causes). But, the conditional dependency struc-
ture will contain conditional dependencies for every causal relation in the causal graph. In
this sense, the conditional independence structure can generate possible hypotheses about
causal paths, but cannot be used to infer causal paths directly (see Ryan et al., 2022, for a
more detailed discussion of the problems of causal inference from network models). But, for
those who do want to take a next step and identify directed causal graphs, causal discovery
is an exciting field with many advances, such as causal discovery algorithms that do not
require the absence of unobserved common causes or feedback loops (Eberhardt, 2017).
There are at least three reasons for why one might want to model the conditional
independence structure of the MRF rather than going a step further and using the MRF
to discover directed causal graphs. First, inferring a DAG from conditional dependencies
in observational data requires strong assumptions that may not hold in practice (e.g., no
unobserved common causes and no feedback loops). Second, for a conditional independence
structure, there may be many directed causal graphs that are equivalent and consistent
with the conditional independence structure. We have already seen that there are several
equivalent graphs for the three-variable example above, and for more than three variables
the set of equivalent graphs increases enormously. Therefore, it may be much easier to work
with a single MRF than with the potentially large set of equivalent causal graphs (Epskamp,
Borsboom, & Fried, 2018). Third, the MRF does not commit one to a causal interpretation;
instead, one can choose a purely statistical interpretation of predicting variables from other
variables in the network or other interpretations (e.g., Epskamp et al., 2022).

Bayesian Graphical Modeling

Bayesian inference aims to use data to update our knowledge about the network
structure Ss —the collection of edges in the network— and the network parameters Θs
—the edge weights. To allow the data to update our knowledge of the network structure
and parameters, we need to make explicit what we know about them before seeing the
data. Figure 1 shows that there are many possible structures that that could underlie the
network, and similarly, there are many possible values for the corresponding edge weights.
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 7

But which values could describe our data? Since our goal is to learn about them, the
specific configuration of the network relations and the exact parameter values are usually
unknown to us. To account for this uncertainty, we assign prior distributions to the model
or structure Ss and to the parameters of that model Θs . A prior is a probability distribution
that a Bayesian uses to assign weights (i.e., probability or probability density) to different
values of the parameters and structure. First, we assign prior probabilities to the different
network structures p(Ss ) (i.e., the prior distribution of the effect), and then, conditional
on a particular structure, we specify prior distributions on the corresponding edge weights
p(Θs | Ss ) (i.e., the prior distribution of the effect size). The priors provide a way to
formalize theory and incorporate advanced knowledge (e.g., results from previous research;
Lindley, 2004; Vanpaemel & Lee, 2012), or they can be used to express ignorance using
a default or objective prior specification (e.g., Consonni, Fouskakis, Liseo, & Ntzoufras,
2018). In the appendix, we provide details about the prior distributions implemented in
three popular R packages for analyzing MRF models.
Regardless of how we specify the priors, Bayes’ rule weighs the prior distribution with
the information coming from the observed data to update it to a posterior distribution,

p(data | Θs , Ss ) p(Θs | Ss ) p(Ss )

p(Θs , Ss | data) = .
p(data)

This joint posterior distribution expresses everything that we know about the structure
and parameter values of the network after seeing the data and is central to the Bayesian
analysis of graphical models. The different Bayesian tests for conditional independence
consider different aspects of this joint posterior. To make this more explicit, we factor the
joint posterior as follows

p(Θs , Ss | data) = p(Θs | Ss , data) × p(Ss | data),

and express it as a product of the posterior distribution of the parameters Θs under the
specific structure Ss , and the posterior distribution of the possible structures with the
parameters integrated out. The former is referred to as the conditional posterior distribution
for the network parameters (i.e., it is the posterior distribution of the edge weights for a
specific structure Ss ) and the latter as the marginal posterior distribution of the network
structure (i.e., a posterior of the structures without the parameter values for the edge
weights). Below, we will use the conditional posterior distribution p(Θs | Ss , data) for
Bayesian parameter estimation, and the marginal posterior distribution p(Ss | data) for
Bayesian hypothesis testing.

Bayesian Hypothesis Testing: The Bayes Factor

Two out of the three proposed Bayesian methods for testing the conditional indepen-
dence hypothesis that we review in the next section make use of the Bayes factor (Jeffreys,
1939; Kass & Raftery, 1995). The Bayes factor quantifies the relative predictive perfor-
mance of two rival hypotheses (e.g., the conditional dependence of two variables or their
conditional independence), or of two competing models or structures. Consider two com-
peting network structures Ss and St . The Bayes factor is defined as the change in beliefs
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 8

concerning the relative plausibility of the two structures before and after observing the data

p(Ss ) p(data | Ss ) p(Ss | data)

× = .
p(St ) p(data | St ) p(St | data)
| {z } | {z } | {z }
Prior BFst Posterior
odds odds

Specifically, the first factor on the left of the formula above is the prior odds, that is, the
relative plausibility of the two structures before having seen the data. The second factor is
the Bayes factor which indicates the statistical evidence or support for the two structures in
the data at hand. The term on the right is the posterior odds, which indicates the relative
plausibility of the rival models after having seen the data. In this paper, we assume that
the prior odds are equal to one by assuming p(Ss ) = p(St ), which makes the Bayes factor
equal to the posterior odds (see Marsman, Huth, Waldorp, & Ntzoufras, 2022, for a different
approach).
The subscripts in the Bayes factor notation indicate in which direction the support
is expressed. BFst indicates the relative support for Ss over St and BFts indicates the
relative support for St over Ss . Observe that the Bayes factor BFts is the reciprocal of
BFst , i.e., BFts = 1/BFst . The Bayes factor BFst ranges from 0 to ∞, values larger than
one indicate a relative support for Ss while values smaller than one indicate the relative
support for St . If the Bayes factor is equal to one, both structures predicted the data equally
well. In practice, we usually interpret Bayes factors between 1/10 and 10 as evidence that is
insufficiently compelling.3 .

Bayesian Estimation: The Posterior Distribution of Partial Associations

One of the three Bayesian methods for testing the conditional independence hypoth-
esis that we review in the next section makes use of Bayesian estimation. In practical
situations, we often wish to estimate the parameters Θ′ for a particular structure S ′ . This
could be the structure with the highest posterior probability p(S ′ | data), the median prob-
ability structure (e.g., Barbieri & Berger, 2004; Marsman et al., 2022) that consists of all
relations for which the posterior inclusion probability (defined in the section on Bayesian
model averaging) is greater than a half, or it could be the complete structure that includes
all relations. The posterior distribution for a single structure S ′ is

p(data | Θ′ , S ′ ) × p(Θ′ | S ′ )
p(Θ′ | S ′ , data) = ,
p(data | S ′ )
where p(Θ′ | S ′ ) denotes the prior distribution for the parameters under the structure S ′ .
Given a single parameter θij , the prior p(θij′ | S ′ ) assigns a relative plausibility to each value

of the parameter. The information in the data is then used to update this prior distribution
into a posterior distribution p(θij | S ′ , data). In the posterior distribution, the plausibility
of parameter values that predict the data well increases, while the plausibility of parameter
values that predict the data poorly decreases (Wagenmakers et al., 2016).
3
In principle, Bayes factors are a continuous measure of evidence and therefore do not require strict cut-
off values. But even if we do, there is no hard and fast rule for what the cutoff should be, and practitioners
may prefer other values (Jeffreys, 1961; Kass & Raftery, 1995).
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 9

Instead of reporting the full posterior distribution for each element in Θ′ , we often
report it in terms of a measure of location (i.e., the posterior mean, median, or mode)
and spread (i.e., the posterior variance), or in terms of an x% credible interval (see van
Doorn et al., 2021). An x% credible interval contains x% of the probability mass of the
posterior distribution. Two popular ways to create an x% credible interval are the highest
posterior density interval, which is the shortest possible credible interval that contains x%
of the posterior mass, and the x% central credible interval, which is obtained by clipping
(100−x)/2% from each tail of the posterior distribution. Figure 2 shows a fictional example of
a posterior distribution that has a 95% central credible interval and a 95% highest density
interval. The posterior is a probability density with the gray area under its curve containing
95% of its total probability. Note that the highest density interval is shorter than the central
credible interval, even though both capture 95% of the posterior.

Univariate Posterior Distribution for θ

Posterior Density

ROPE

95% HDI

95% Central CRI

Figure 2 . An example of a posterior distribution for a parameter θ, the line at the bottom
of the density represents the 95% central credible interval and the shaded gray region
represents the 95% highest density interval (HDI). The two dashed vertical lines around
zero represent the region of practical equivalence (ROPE, introduced in the next section).

Equipped with these Bayesian concepts, we next turn to the three proposed Bayesian
approaches for testing conditional independence.

Three Bayesian Methods for Testing Conditional Independence

Approach 1: Credible Interval

In frequentist statistics, an assessment of whether or not the null value θ0 falls within
the x% confidence interval for a parameter θij (sometimes considered as estimation, but see
Morey, Hoekstra, Rouder, Lee, & Wagenmakers, 2016) is equivalent to the test of the null
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 10

hypothesis,
H0 : θij = θ0 ,
with a significance level of α = (100 − x)%: We would reject H0 with significance level α if
the null value falls outside the (100 − α)% confidence interval (cf. Figure 2). It is tempting
to extend this testing approach to Bayesian statistics by using an x% credible interval to
test whether or not we could reject H0 . But from which posterior distribution should we
take the credible interval for the partial associations? In practice, this is usually done
using a complete structure SC that includes all relations (e.g., the bottom right structure
in Figure 1). However, this approach implies that the relation between nodes i and j is a
priori assumed to exist, and we are thus testing a hypothesis that we assume to be false
from the outset (e.g., Jeffreys, 1939). This signals a bias against the null hypothesis, which
is common in classical null hypothesis significance tests.
In practice, the logic behind credible interval-based tests may indeed lead to contra-
dictions, since comparisons between the null hypothesis and its complement using the Bayes
factor, for example, may signal support for the null hypothesis of conditional independence,
while the null value θ0 would fall outside the credible interval. See Berger and Delampady
(1987) and Wagenmakers, Lee, Rouder, and Morey (2020) for detailed discussions of this
issue. Null hypothesis tests based on the credible interval can also lead to ambiguous re-
sults, because if the null value would fall within the interval, we cannot interpret this as
support for the null hypothesis because the test cannot distinguish between the potential
causes of this failure to reject (i.e., absence of evidence or evidence of absence). In order to
test for conditional independence, we must therefore be able to quantify support in favor of
the null hypothesis.
Despite this complication, credible interval-based tests have been used to test for con-
ditional independence in the Bayesian graphical modeling literature. For example, Jonger-
ling, Epskamp, and Williams (2022) use credible intervals to perform edge selection (i.e.,
conditional independence testing) in GGMs with the goal of estimating the posterior distri-
bution of centrality measures. Williams (2021) used a generalization of the credible interval
test based on the idea that we can specify a region in the parameter space that is essentially
zero —the region of practical equivalence (ROPE, Kruschke, 2011)— and then exclude an
edge if x% of the posterior distribution of the partial association is inside the ROPE, oth-
erwise include it (cf. Figure 2). In a slightly different way, Marsman et al. (2022) also used
credible intervals for edge selection. They used a continuous spike and slab prior on the par-
tial associations of an Ising model, where the intersection of the spike and slab components
occurs at an approximate x% credible interval. This is very similar to using ROPE for edge
selection; to set the spike-and-slab prior, Marsman et al. (2022) also start with a posterior
distribution that assumes the effect is present (based on the unit-information prior; Kass &
Wasserman, 1995). However, unlike the credible interval test and the ROPE approach, the
approach in Marsman et al. (2022) can distinguish the potential causes underlying the edge
exclusion because it assigns prior weights to the edge inclusion and exclusion hypotheses.
Thus, our concerns with credible interval-based tests are directed at their conceptual
underpinnings, particularly their inability to quantify support for the null hypothesis. To
quantify this support, we need an evidential measure that contrasts the competing hypothe-
ses of conditional dependence and independence, i.e., the Bayes factor. For a review of (log)
Bayes factors as weight of evidence see, for example, Good (1985).
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 11

Approach 2: The Single-Model Bayes Factor

The Bayes factor is the gold standard for Bayesian hypothesis testing (Berger & Peric-
chi, 2015), and around the same time that graphical models became popular in psychology,
Bayes factor hypothesis testing became popular in psychological research. In large part,
this increased popularity of the Bayes factor in psychology is a response to the misuse of
the null hypothesis significance test (NHST) in psychological research and the limited repli-
cability (Ioannidis, 2005; Open Science Foundation, 2015) of many psychological findings
established with NHST (e.g., Wagenmakers, 2007; Wagenmakers, Wetzels, Borsboom, &
van der Maas, 2011). Some of the concerns that methodologists have with NHST also play
a role in the credible interval test of the previous section. One of the more prominent
concerns is that the adequacy or inadequacy of the null hypothesis is not compared against
an alternative. Thus, rejection of the null hypothesis should not be taken as evidence in
favor of the alternative hypothesis, which may be just as inadequate as the null hypothesis
(or even more inadequate). The Bayes factor, however, compares the predictive adequacy
of the null hypothesis against that of an alternative, and as such can separate evidence
for the absence of an effect, evidence for the presence of an effect, but also the absence of
evidence in either direction (e.g., Dienes, 2014; Keysers, Gazzola, & Wagenmakers, 2020).
Thus, Bayes factor testing is a significant step forward for psychological network analysis.
However, we are concerned with the way it is commonly formulated.
We consider the Bayes factor test for the conditional independence of the variables i
and j in the network, i.e., we consider the following two hypotheses

H0 : θij = 0, and H1 : θij ̸= 0.

If we wish to assign prior probabilities to these hypotheses, it is easier to reformulate them in

terms of the edge indicators and model the size of the effect θij conditional on the presence
of the effect. That is, we model p(Θs | Ss ) such that we can impose a prior on the hypothesis
or model Ss . Then, our hypotheses can be reformulated as

H0 : γij = 0, and H1 : γij = 1.

By formulating the hypothesis in terms of the edge indicator rather than the edge weight, we
immediately encounter a problem. We cannot yet isolate the effect of a single relationship,
i.e., the edge indicator, and thus we must now carefully consider how to set up the Bayes
factor. The way this is usually done is by comparing two structures Ss and St that are
identical except that the relation between the variables i and j is present in Ss but is absent
in St . In this way, comparing Ss with St using the Bayes factor gives us a Bayes factor
test for H1 versus H0 . Although it is not made explicit, in practice the complete structure
Ss = SC is used here, as in the case of the credible interval test we discussed earlier.
Note that the Bayes factor test for conditional independence we formulated above is
not uniquely defined. In principle, we could compare any two structures Ss and St , as long as
they are identical except that the relation between the variables i and j is present in Ss but
not in St . For our hypothetical three-variable example, this means that we have three ways
to test the conditional independence of variables A and I: We could contrast Ss = S3 with
St = S1 , Ss = S5 versus St = S2 , or Ss = S8 versus St = S6 . Each of these comparisons is a
valid comparison in terms of contrasting the effect of the relation (i.e., assessing conditional
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 12

independence). However, in each case we are making a different assumption about the other
relationships in the network. We will refer to any such Bayes factor test as a single-model
Bayes factor, since it assumes a single model for the remaining relationships in the network.
The single-model Bayes factor test is sensitive to the assumption concerning the overall
network structure because partial associations are sensitive to the other partial associations
in the model or structure. To illustrate, consider the relation between variables A and I in
our three-variable example. First, in order to compute its value in our example, we express
the Bayes factor as a function of the prior and posterior probabilities:

p(Ss | data) p(Ss )


BFst = .
p(St | data) p(St )
| {z } | {z }
Posterior Prior
odds odds

The posterior probabilities for each of the eight structures are shown in Figure 1. When we
assume that each of the structures is equally plausible a priori, the prior probabilities are
equal to one and the Bayes factors are equal to the posterior probabilities. Thus, the Bayes
factors for the three possible model pairs are obtained as follows:

p(S3 | data) .016

BF31 = = = 0.23,
p(S1 | data) .07
p(S5 | data) .019
BF52 = = = 0.17,
p(S2 | data) .11
p(S8 | data) .03
BF86 = = = 0.05.
p(S6 | data) .66

This demonstration confirms that the single-model Bayes factor can, in fact, be sen-
sitive to our choice for the remaining relations in the network. The first two Bayes factors
(i.e., BF31 = 0.23 and BF52 = 0.17) showed weak evidence for exclusion, while the third
Bayes factor showed strong evidence for exclusion (i.e., BF86 = 0.05). But which Bayes
factor test should we use?
Williams and Mulder (2020a) proposed the single-model Bayes factor for testing con-
ditional independence in GGMs (see also Giudici, 1995). In their approach, the complete
structure is used as a basis for comparison. In the next section, we show that this method
works well when the data generating structure has relatively many relations, consistent with
the model’s assumption, but it starts to perform less well when the data generating struc-
ture is sparse and has relatively few connections. Since we are typically highly uncertain
about which particular structure would underlie our data (see Marsman & Haslbeck, 2023;
Marsman et al., 2022), the foundations of the single-model Bayes factor can be unstable.

Approach 3: The Inclusion Bayes Factor

We can use Bayesian model averaging (BMA; Hoeting et al., 1999; Kaplan, 2021)
to overcome the sensitivity of the single-model Bayes factor to our assumptions about the
remaining relationships in the network. When we consider the single-model Bayes factor, we
must assume that the network is based on some structure. In practice, however, we usually
do not know what that structure is. To account for our uncertainty about the structure
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 13

of the network, BMA considers all possible structures and weights the outcome of each
structure by its posterior probability; the relative plausibility that the structure produced
the data at hand. By weighting the outcome of each structure by its posterior probability,
BMA accounts for the uncertainty we have about which structure is at play (Hinne, Gronau,
van den Bergh, & Wagenmakers, 2020; Huth, de Ron, et al., 2023). A. Mohammadi and
Wit (2015) and Marsman et al. (2022) applied BMA to graphical models.
We focus here on the posterior inclusion probability, the posterior probability of
including an effect, which we use to estimate the inclusion Bayes factor; the Bayes factor
test that pits the conditional dependence hypothesis against the conditional independence
hypothesis. Although we do not consider it here, BMA is also useful for estimating the
marginal posterior distribution for the partial associations; a robust estimate of the effect
size that incorporates the uncertainty in the parameter and the uncertainty in its selection.
We can express the posterior probability of including the edge between variables i
and j as the sum of the posterior probabilities over all structures that include the edge. Let
S (i−j) denote the set of structures that include an edge between variables i and j, then the
inclusion probability can be computed as

p(S ′ | data),
X
p(γij = 1 | data) =
S ′ ∈S (i−j)

which weights the posterior plausibility of the inclusion of the relation in the network
structure. For example, the posterior inclusion probability of including the relation between
variables A and I (i.e., γAI = 1) in Figure 1 is equal to

p(γAI = 1 | data) = p(S3 | data) + p(S5 | data) + p(S7 | data) + p(S8 | data)
= .016 + .019 + .005 + .03 = .07.

Since the posterior probabilities for edge inclusion and exclusion sum to one, the corre-
sponding probability of exclusion is p(γAI = 0 | data) = 1 − p(γAI = 1 | data) = .93. The
Bayes factor for inclusion can now be determined as follows (Huth, de Ron, et al., 2023;
Marsman & Haslbeck, 2023; Marsman et al., 2022)

p(data | γij = 1) p(γij = 1 | data) p(γij = 1)


= .
p(data | γij = 0) p(γij = 0 | data) p(γij = 0)
| {z } | {z } | {z }
Inclusion Posterior Prior
Bayes factor inclusion odds inclusion odds

The inclusion Bayes factor quantifies the weighted evidence for the inclusion of the relation-
ship across all structures. As such, the inclusion Bayes factor provides a simple measure to
distinguish between inconclusive evidence and conclusive conditional independence between
two nodes. When we assume that all structures are equally likely a priori, the prior inclu-
sion probability for individual edges is equal to 1/2. The prior odds then equal to 1, and we
see that the inclusion Bayes factor for including the edge between variables A and I is equal
to .07/.93 ≈ .074, which means that based on the information in our data, we have strong
evidence that an edge between variables A and I should be excluded from the network,
in other words, we have strong evidence for conditional independence (i.e., the exclusion
Bayes factor is 1/.074 ≈ 13.5). Note that the inclusion Bayes factor does not depend on
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 14

the remaining relationships in the network, since it averages the network structures and
thus overcomes the dependence of the single-model Bayes factor on assumptions about the
remaining relationships.

Simulation Study
We performed a simulation study to compare the accuracy of edge selection using the
three methods in the case of a GGM using the BDgraph R package (R. Mohammadi & Wit,
2019). The R code we used in our simulations is available in the repository at https://
osf.io/2x74v/. We simulated several conditions. Specifically, we varied the size of the
network, p = {10, 30, 50}, the number of observations, n = {100, 200, 500, 1,000, 5,000},
the size of the focal edge weight between variables 1 and 2 (i.e., partial correlation), θ12 =
{0, .1, .25, .4}, and the density of the rest of the network (i.e., the number of relations in
the rest of the network). We simulated the structures based on a random graph. We varied
the density of the network so that the probability of an edge between two nodes was either
.2, .5, or .8. Given the generated structure, we sampled the remaining edge weights from
a g-Wishart distribution (Roverato, 2002). Since manipulating the edge weight between
variables 1 and 2 could result in a precision matrix that is not positive semi definite we
continued sampling precision matrices until we found one that was positive semi definite.
We obtain the single-model (non-BMA) parameters, by sampling from their poste-
rior distribution of the edge weights based on the full model. In this case this posterior
distribution is a g-Wishart distribution (Lenkoski, 2013; Roverato, 2002). We obtain the
single-model Bayes factors by computing the fraction of the normalizing constants (i.e.,
marginal likelihoods) of the g-Wishart distributions under the fully connected structure
and a structure that excludes the focal edge. For BMA analysis we used the default set-
tings from the function bdgraph. For each dataset, using 10,000 iterations4 , we computed
for the focal edge weight θ12 :

1. The central credible interval for the single-model parameter estimate, and whether or
not it included the test-relevant value of 0. We then transformed this into a quasi-
inclusion probability, which was 1 if the interval included 0 and 0 otherwise, to make it
comparable to the other measure. We computed two variants of the credible interval:
(i) the standard 95% central credible interval and (ii) an adaptive credible interval.
The latter is equivalent to ROPE (for more details, see, Kruschke, 2011).

2. The single-model posterior edge inclusion probability. The single-model posterior

inclusion probability is calculated from the single-model Bayes factor as follows

BF10 O10
p(γij = 1 | data) = ,
1 + BF10 O10
where BF10 is the single-model Bayes factor in favor of conditional dependence and
O10 is the prior odds. We assumed full structure for the remaining relationships in
the network and used O10 = 1 in our analysis.
4
Note that we ran the MCMC procedures for a fixed number of iterations, and did not check for conver-
gence of the individual Markov chains. Although our experience is that the implemented procedures tend
to converge quickly, there is no guarantee that the chains that were used in our simulations actually did.
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 15

3. The posterior edge inclusion probability obtained from Bayesian model averaging.

We computed the Brier score (Brier, 1950), which quantifies the mean squared differ-
ence between predicted probabilities and actual outcomes for a binary event (in this case
the presence of an edge), with lower scores indicating better predictive performance. For
each metric and condition Figure 3 shows that when the focal parameter has an edge weight
equal to zero (i.e., conditional independence), the inclusion Bayes factor and the adaptive
credible interval perform best across different sample sizes, numbers of variables, and net-
work densities. We can also observe that the performance of the single-model Bayes factor
becomes worse as the network density and the number of variables increase. All methods
tend to perform better as a function of sample size. When the edge is present, even with
a value of θ12 = 0.1, we see that the situation is reversed, in other words, the 95 % central
credible interval and the single-model Bayes factor perform better than the inclusion Bayes
factor and the adaptive credible interval, especially for N < 1,000. When the value of the
partial correlation is 0.25 and 0.4, all of the methods tend to perform quite well.

Partial Correlation: 0 Partial Correlation: 0.1 Partial Correlation: 0.25 Partial Correlation: 0.4
1.0

D: 0.2
p: 10
0.5

0.0
1.0

D: 0.5
p: 10
0.5

0.0
1.0

D: 0.8
p: 10
0.5

0.0
1.0

D: 0.2
p: 30
Average Brier Score

0.5

0.0
1.0 SM−BF
D: 0.5
p: 30
0.5 Inclusion−BF
95% CRI
0.0 Adaptive CRI
1.0
D: 0.8
p: 30

0.5

0.0
1.0
D: 0.2
p: 50

0.5

0.0
1.0
D: 0.5
p: 50

0.5

0.0
1.0
D: 0.8
p: 50

0.5

0.0
100 200 500 10005000 100 200 500 10005000 100 200 500 10005000 100 200 500 10005000
Sample size
Figure 3 . Average Brier score for each of the four measures as a function of the sample size
plotted for each value of the edge weight, number of variables (p), and network density (D).

Figure 3 shows that the density of the network has an influence on which method
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 16

performs best. To get a clearer picture of the overall performance, we first aggregate the
accuracy of the methods across effect sizes and compute the values for the area under the
receiver operating characteristic curve (AUC) for each measure. The receiver operating
characteristic (ROC) curve plots the trade-off between the true positive rate (sensitivity)
and the false positive rate (1 - specificity) as we vary the classification threshold. Therefore,
the AUC is a performance measure of how well the methods can capture the truth — in this
case, whether the edge is truly present. Methods with a higher AUC value (closer to 1) can
better discriminate between present and absent edges than methods with lower AUC values
(see Fawcett, 2006, for an introduction to ROC curves and AUC values). As can be seen
from the results presented in Figure 4, the inclusion Bayes factor performs better than the
single-model Bayes factor for low and medium network density levels, especially for smaller
sample sizes, but performs worse when the network density is high. When the density is
high, the structure assumed by the single-model Bayes factor is close to the true underlying
network structure (i.e., both are densely connected), and thus the single-model Bayes factor
has an advantage under this condition. The BMA approach still assumes different structures
for the data and is therefore suboptimal when the true structure is dense. The 95 % credible
interval shows the worst performance overall.
Since the two Bayes factor approaches are the only formal ways to test for conditional
independence hypotheses, we wish to compare their performance in some more detail. Figure
5 plots the proportion of times the Bayes factors made a correct decision in detecting
evidence for the true hypothesis. As can be seen, and as expected based on the previous
plots, when the edge is absent (i.e., when θ12 = 0), the inclusion Bayes factor outperforms
the single-model Bayes factor in all simulation conditions. This suggests that the inclusion
Bayes factor is quite good at capturing evidence in favor of conditional independence. When
the edge is present, its weight is small, and the network is small and sparse, the two Bayes
factors show similar performance. In contrast, as can also be seen in Figure 4, as the
network becomes larger and more densely connected, the single-model Bayes factor begins
to outperform the inclusion Bayes factor. As the true value for the edge weight increases,
both methods perform very well, especially for large sample sizes.

Empirical Example
To illustrate the difference between the two Bayes factors we consider the analysis of
a data set from a study by Gojković, Dostanić, and Ðurić (2022) on the network structure of
empathy, narcissism, and the Dark Triad (i.e., the combination of narcissism, psychopathy,
and Machiavellianism) personality traits. The data are publicly available at https://
osf.io/7jcks/. It consists of eight variables, each measured by a battery of Likert-scale
items. The narcissism, psychopathy, and Machiavellianism variables are based on the 27
items from the Short Dark Triad (i.e, each variable is a sum of responses to 9 different
items Jones & Paulhus, 2014); the cognitive empathy, affective resonance, and affective
dissonance variables are based on the 36 items from the Affective and Cognitive Measure
of Empathy (Vachon & Lynam, 2016); the narcissistic admiration and narcissistic rivalry
variables are based on the 18 items from the Narcissistic admiration (Adm)iration and
Narcissistic Rivalry (Back et al., 2013). The affective dissonance items were inversely coded
so that a higher summed score corresponded to a higher level of affective dissonance. The
study was based on a sample of 263 high school and university students from Vojvodina,
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 17

Density: 0.2 Density: 0.5 Density: 0.8

1.00

0.95

0.90

p: 10
0.85

0.80

0.75

0.70
1.00

0.95

0.90
SM−BF
AUC

p: 30
0.85 Inclusion−BF
95% CRI
0.80 Adaptive CRI

0.75

0.70
1.00

0.95

0.90

p: 50
0.85

0.80

0.75

0.70
100 200 500 1000 5000 100 200 500 1000 5000 100 200 500 1000 5000
Sample size
Figure 4 . AUC values as a function of the sample size plotted for different values of the
network density and number of variables p.

Serbia.
Since we wish to see if there is a difference in the conclusion we would draw from using
the two Bayes factors, we analyzed the network structure of the eight variables with a GGM
using both the single-model and multi-model or BMA perspectives. For the single-model
analysis, we estimated the parameters of a fully connected GGM by drawing one million
samples from the corresponding posterior distribution, which in this case is a g-Wishart
distribution (Lenkoski, 2013; Roverato, 2002). The Bayes factor was computed for each of
the 8×(8−1)/2 = 28 edges in the network by computing the ratio of the marginal likelihood
with all edges present to the marginal likelihood with the focal relationship excluded. We
used the BDgraph package to sample from the g-Wishart distribution and to compute the
marginal likelihood. For the multi-model analysis, we also used the BDgraph package, which
estimates the posterior inclusion probabilities using a Markov chain Monte Carlo procedure.
We used one million iterations for each Markov chain. In each of these analyses, we used
the default settings of BDgraph, setting a g-Wishart prior on the precision matrix Θ and
assuming a prior inclusion probability of 1/2 for all edges.
Figure 6 illustrates that there is indeed a difference between the inferences we would
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 18

Partial Correlation: 0 Partial Correlation: 0.1 Partial Correlation: 0.25 Partial Correlation: 0.4
Proportion of Times Evidence was found in Favor of the True Hypothesis (BF > 10)
1.0

D: 0.2
p: 10
0.5

0.0
1.0

D: 0.5
p: 10
0.5

0.0
1.0

D: 0.8
p: 10
0.5

0.0
1.0

D: 0.2
p: 30
0.5

0.0
1.0

D: 0.5
p: 30
0.5 SM−BF
Inclusion−BF
0.0
1.0

D: 0.8
p: 30
0.5

0.0
1.0

D: 0.2
p: 50
0.5

0.0
1.0

D: 0.5
p: 50
0.5

0.0
1.0

D: 0.8
p: 50
0.5

0.0
100 200 500 1000 5000 100 200 500 1000 5000 100 200 500 1000 5000 100 200 500 1000 5000
Sample size
Figure 5 . The proportion of times that the two Bayes factors found evidence in favor of the
true hypothesis, as a function of the sample size plotted for each value of the edge weight,
number of variables (p), and network density (D).

draw using the inclusion Bayes factor and the single-model Bayes factor. We can see that the
inclusion Bayes factor provides evidence for edge exclusion, i.e., for the estimated parameters
that are close to zero, as indicated by the narrower "v" shape shown in the left panel. In
the BMA case, there is a more pronounced shrinkage towards zero. Therefore, as shown in
the previous section, the inclusion Bayes factor offers more pronounced evidence in support
of conditional independence than the single-model Bayes factor.
Figure 7 shows the edge evidence plots — networks whose edges reflect strong evidence
for edge inclusion (using a cutoff of BF10 = 10). Based on the inclusion Bayes factor in
the left panel, we conclude that 13 of the 28 possible edges are present in the network, and
based on the single-model Bayes factor in the right panel, we conclude that 12 of them are
present. For the edge between the variables psychopathy (SD3P) and affective resonance
(ARe), the exclusion Bayes factor is equal to BF01 = 9.1, close to the evidential cutoff of 10,
giving us evidence in favor of conditional independence. For comparison, the largest single-
model Bayes factor in favor of edge exclusion is between the variables admiration (Adm) and
 BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 19

Inclusion Bayes factor Single−Model Bayes factor

2
log(BF10)

−2

−0.4 −0.2 0.0 0.2 0.4 0.6 −0.4 −0.2 0.0 0.2 0.4 0.6

θ θ

Figure 6 . The (natural) logarithm of the Bayes factors plotted against the posterior mean
of the corresponding edge weight. The left panel shows the results for the inclusion Bayes
factor, and the right panel shows the results for the single-model Bayes factor. Bayes
factor values greater than or equal to one hundred are set equal to one hundred (i.e.,
log(BF10 ) = 4.6.

affective resonance (ARe), and is only BF01 = 2.5. Examining the networks in Figure 7, we
can see that, for example, with the inclusion Bayes factor we find evidence for the inclusion
of an edge between the variables psychopathy (SD3P) and admiration (SD3N), cognitive
empathy (CEm) and admiration (SD3N), but we have no evidence for the inclusion of the
same edges when we use the single-model Bayes factor. Conversely, using the single-model
Bayes factor, we find evidence for the inclusion of an edge between the variables cognitive
empathy (CEm) and admiration (Adm), for which we have inconclusive evidence when
using the inclusion Bayes factor. From our simulations, we know that When the network
structure is sparse, which appears to be the case in this example, the inclusion Bayes factor
can more accurately capture the evidence, both for edge inclusion and edge exclusion.
Since we argue that the credible and/or highest density intervals should not be used
for hypothesis testing, we adhere to this principle in this section. However, because these
intervals are valuable measures of posterior parameter uncertainty, we present plots of the
95% central credible intervals around each posterior edge weight. We computed the 95%
central credible intervals for the BMA parameter estimates and the 95% central credible
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 20

Single-Model Bayes
Inclusion Bayes Factor
Factor

Figure 7 . Edge evidence plots based on the inclusion Bayes factor on the left and the
single-model Bayes factor on the right. The blue solid lines indicate edges for which there is
a BF10 ≥ 10, the dashed red line indicates an inclusion Bayes factor that almost reaches the
exclusion threshold and the dashed grey lines indicate edges for which there is inconclusive
evidence for edge (in)exclusion.

intervals for the posterior parameter estimates based on a structure that assumes all edges
are present. As can be seen in Figure 8, the central credible intervals obtained from the
two methods are different. We prefer the credible intervals based on BMA because they
account for both parameter uncertainty and structure uncertainty.

Discussion

In this paper, we have reviewed three different Bayesian approaches to testing condi-
tional independence hypotheses for a class of Markov random field models used in network
psychometrics. The first method uses the posterior distribution of the partial association θij
to check whether it falls in the ROPE, or similarly whether its x% credible interval contains
zero. Both scenarios would indicate that the hypothesis of conditional independence of the
variables i and j cannot be rejected, but the drawback is that we cannot use it to support
the independence hypothesis. The second approach used the single-model Bayes factor to
test for conditional independence, which compares two network structures Ss and St that
are identical except that the focal relationship is included in Ss but not in St . Although
this method could be used to express support for the conditional independence hypothesis,
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 21

Credible Intervals

BMA
Single-
Model

Figure 8 . The 95 % credible intervals for the BMA estimates in black and for the estimates
based on a fully connected structure in yellow. The vertical dotted line represents the value
of θ = 0. The points on each line represent the posterior median estimates.

its drawback is that it is sensitive to the required choice of which relations are in the rest
of the network. The third approach uses BMA to express the inclusion Bayes factor, which
accounts for the uncertainty about other relations in the network. The inclusion Bayes
factor is free from the conceptual problems of credible interval-based tests and is optimal
when we are uncertain about the structure underlying our data.
In the simulations, we showed that the inclusion Bayes factor was the best overall
method for determining conditional independence. It also performed well in determining
conditional dependence, although the single-model Bayes factor outperformed the inclusion
Bayes factor in scenarios where the true network structure is densely connected. In these
scenarios, which are close to the assumption of a fully connected structure underlying the
single model Bayes factor, the inclusion Bayes factor loses power because it continues to
consider alternative structures for the data at hand. However, in practice, since we do not
know what the underlying structure is, the inclusion Bayes factor is the most robust choice
for inferring conditional independence or dependence.

Critique: The correct model is probably not being considered

The mathematics behind Bayesian model comparison does not assume that any of the
models under consideration are correct in some abstract sense, as the formulas only evaluate
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 22

the predictive adequacy of the models under consideration (see for instance, O’Hagan, 2010,
p. 167). Nevertheless, many statisticians have argued that Bayesian model comparison
only makes sense if the correct model is in the collection of models under consideration
— the M-closed context (Bernardo & Smith, 1994, pp. 383–407). The main concern
of critics of Bayesian model comparison, and BMA in particular, is that the posterior
distribution cannot converge to the correct model if it is not in the collection of models
under consideration — the M-open context. Instead of converging to the correct model,
the posterior distribution would converge to the model that is closest to the true model in
a Kullback-Leibler sense in the M-open context. This model would be optimal in terms of
its predictive adequacy relative to the collection of models under consideration.
Box’s famous adage “all models are wrong” (Box, 1976, p. 792) is often used to make
the case that the M-closed assumption is also wrong. There are two ways in which we think
the true model might differ from the one we consider in psychological network modeling.
First, the network models we use typically include main effects and pairwise relations (i.e.,
first and second-order interactions). In principle, one could consider models with third or
higher-order interactions, but these models are computationally demanding. Second, we
often have a substantive motivation for choosing the variables to include in our network,
but this choice can have a huge impact on the network structure. For example, two variables
will be conditionally dependent if we exclude their common cause from the network, but
conditionally independent if we include it. This is called the boundary specification problem
(Laumann, Marsden, & Prensky, 1989; Neal & Neal, 2023). However, it is likely that if we
knew which variable caused other variables, we would include it in the network. Thus,
while we agree that the M-closed assumption is unlikely to hold in practice, we also agree
with the continuation of the adage that “all models are wrong, but some models are useful”
(Box & Draper, 1987, p. 424). With BMA, we evaluate the predictive adequacy of the
structures of interpretable network models formulated on a substantively interesting subset
of variables.

Limitation: There are no substantively motivated or good default prior distri-

butions for psychological networks
The BMA approach requires us to specify our prior knowledge and expectations about
the structure of the network. However, despite the large body of literature on psychological
network modeling, we still have a limited understanding of their structure. The main reason
for this limited understanding is that Bayes factor tests that can quantify the support
for certain relational patterns have only recently been proposed and have not yet gained
much traction. As a result, we must rely on standard, objective specifications of priors for
psychometric topologies and their associated parameters. These priors may be inappropriate
for psychological networks for two reasons. First, they may give relatively little weight to
the correct structure. In objective specifications, we usually assign a uniform prior on
the possible structures (cf. appendix). Since there is often a huge collection of possible
structures for the network under consideration, only a tiny fraction of the total probability
is assigned to the correct model. Thus, finding the right model is like looking for a needle
in a haystack. Therefore, it would be helpful to know in advance what kind of model we
are looking for. Second, objective priors are often developed in the context of regression
models, and we are unsure if these specifications make sense in the network context. For
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 23

example, in regression, it makes sense to find a sparse collection of variables that can
make accurate predictions. This is because we wish to choose the least complex model
(i.e., the model with the fewest number of predictors) that best predicts new data. But,
in the context of MRF models an absent edge carries a strong assumption, namely of
conditional independence, which indicates that we should not exclude edges by default.
Although the objective priors we use here assign equal probability to including and excluding
individual edges, we need to investigate the suitability of these priors in the network context.
We encourage researchers to always perform sensitivity analyses by estimating the models
under different prior specifications and examining whether and how much the different
specifications alter the conclusions.
In order to advance the specification of good prior densities, we need to advance our
understanding of psychometric network structures. Early discussions about the underlying
structures of psychometric networks were a reaction to the massive popularity of lasso-based
methods, which assume sparse network structures. Alternatives to the lasso have been
proposed that either focus on densely connected networks (e.g., Marsman et al., 2015), or
that aim to strike a balance between sparse and dense network topologies (e.g., Chen, Li,
Liu, & Ying, 2018), but these approaches have not been widely adopted. This means that
we must interpret the sparsity of psychometric networks with caution, especially when data
are limited (Epskamp et al., 2017; Williams, Rhemtulla, Wysocki, & Rast, 2019).
Now that BMA allows us to test our predictions about network topology, we are
entering a new era of network psychometrics. In the next decade, armed with new Bayesian
methodology, we hope to see an advanced understanding of the structure of psychometric
networks, how they differ across measures and populations, and which relationships have
been explained and which have not.

Limitation: There are few BMA methods for analyzing psychological networks

For network researchers to adopt BMA for their analyses, it is imperative that the
methodology be implemented in user-friendly software. Most psychological network model-
ing analyses are performed in the statistical software R, and two R packages now implement
BMA for network analysis. The BDgraph package5 includes methods for analyzing continu-
ous, binary, and ordinal variables (GGMs and latent GGMs; R. Mohammadi & Wit, 2019),
and the bgms package6 for analyzing MRFs of (mixed) binary and ordinal variables (Mars-
man & Haslbeck, 2023). Since most data sets in psychology contain binary and ordinal
variables, these two R packages already cover a lot of ground. The BDgraph package is
now also implemented in the open-source statistical software JASP (see Huth, de Ron, et
al., 2023), which has a graphical user interface that allows users to point and click on their
desired analyses (e.g., Love et al., 2019; Wagenmakers, Love, et al., 2018). The JASP im-
plementation opens BMA-based methods for psychological networks to researchers without
experience programming in R.
Although we argue that Bayes factor approaches, especially the inclusion Bayes fac-
tor, should be preferred for testing conditional independence hypotheses, as shown in the
empirical example, one can still use the credible or highest density intervals around the
5
https://cran.r-project.org/web/packages/BDgraph/index.html
6
https://cran.r-project.org/web/packages/bgms/index.html
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 24

model-averaged parameter estimates (e.g., edge weights) as measures of parameter uncer-

tainty. For these and many other advantages, the interested reader is referred to the newly
developed R package easybgm (Huth, Keetelaar, Sekulovski, van den Bergh, & Marsman,
2023), which allows researchers with less programming experience to use powerful packages
such as bgms and BDgraph to analyze their data and obtain (BMA) Bayes factors as well
as edge uncertainty plots.
The existing software for BMA-based methods for the analysis of psychological net-
works covers several important variable types — e.g., continuous, binary, and ordinal vari-
ables — for cross-sectional applications of networks. However, there are currently no soft-
ware solutions for networks with nominal, discrete, or count variables, or for longitudinal
data designs. The development of BMA methods for analyzing these types of variables and
research designs, and their software implementations, is a fruitful area for future research.

Challenge: Bayesian model averaging can be time consuming

One of the main challenges of BMA is that it must evaluate the collection of models
under consideration. In practice, it is rarely possible to enumerate all possible models, since
the number of structures grows rapidly as the number of variables increases. Therefore, the
R packages that estimate these models rely on Stochastic Search Variable Selection tech-
niques (George & McCulloch, 1993). These techniques are typically implemented through
Markov chain Monte Carlo algorithms (MCMC, see van Ravenzwaaij, Cassey, & Brown,
2018, for an accessible introduction) that iteratively simulate a network structure and its
associated parameters from the joint posterior distribution. As mentioned in the section on
prior distributions, first an edge indicator variable γij is sampled, and then the correspond-
ing edge weight θij is assigned to a particular prior distribution given the sampled value for
the edge indicator. Since the space of possible models is usually large, it is imperative to
run such procedures for enough iterations to sufficiently explore the joint posterior distri-
bution. For some models, such as the GGM, this is usually very fast for the size of data
sets encountered in psychology. However, for binary or ordinal models, MCMC procedures
can take a long time, depending on the sample size. Fortunately, we only need to run the
procedure once to get the full Bayesian benefit.

Conclusion

We have provided a conceptual review of recent Bayesian tests for conditional inde-
pendence of variables in psychological networks. We argued that the two Bayes factor tests
are conceptually superior to frequentist and credible interval-based tests for conditional
independence, in particular because they can express support, or lack thereof, for condi-
tional independence and dependence between the network’s variables. We have shown that
the single-model Bayes factor is sensitive to the assumption that must be made about the
underlying network structure, while the inclusion Bayes factor adequately accounts for the
structure uncertainty. Thus, the inclusion Bayes factor provides researchers with a straight-
forward test of conditional independence and dependence hypotheses. We hope that the
new Bayesian methodology, which focuses on the analysis of the structure of psychological
networks, (i.e., psychometric topology) will help unravel the complex systems underlying
psychological variables.
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 25

Funding. NS, SEK, and MM were supported by the European Union (ERC, BAYESIAN
P-NETS, #101040876). Views and opinions expressed are however those of the author(s) only and
do not necessarily reflect those of the European Union or the European Research Council. Neither
the European Union nor the granting authority can be held responsible for them. KH was supported
by the Centre for Urban Mental Health (University of Amsterdam) and DvdB was supported by
Amsterdam Brain and Cognition (University of Amsterdam).
Acknowledgements. We would like to thank Sacha Epskamp, Joran Jongerling, and one
anonymous reviewer for their constructive feedback on earlier versions of the manuscript.

References

Back, M. D., Küfner, A. C., Dufner, M., Gerlach, T. M., Rauthmann, J. F., & Denissen, J. J. (2013).
Narcissistic admiration and rivalry: Disentangling the bright and dark sides of narcissism.
Journal of Personality and Social Psychology, 105 (6), 1013. doi: 10.1037/a0034431
Barbieri, M. M., & Berger, J. O. (2004). Optimal predictive model selection. Annals of Statistics,
32 (3), 870–897. doi: 10.1214/009053604000000238
Berger, J. O., & Delampady, M. (1987). Testing precise hypotheses. Statistical Science, 2 (3),
317–335. doi: 10.1214/ss/1177013238
Berger, J. O., & Pericchi, L. R. (2015). Bayes Factors. In N. Balakrishnan, T. Colton, B. Everitt,
W. Piegorsch, F. Ruggeri, & J. L. Teugels (Eds.), Wiley StatsRef: Statistics Reference Online.
Wiley. doi: 10.1002/9781118445112.stat00224.pub2
Bernardo, J. M., & Smith, A. F. M. (1994). Bayesian Theory. New York: Wiley.
Blanken, T. F., Isvoranu, A.-M., & Epskamp, S. (2022). Estimating network structures using model
selection. In Network Psychometrics with R (pp. 111–132). Routledge.
Borsboom, D., Deserno, M. K., Rhemtulla, M., Epskamp, S., Fried, E. I., McNally, R. J., . . .
Waldorp, L. J. (2021). Network analysis of multivariate data in psychological science. Nature
Reviews Methods Primers, 1 (1), 58. doi: 10.1038/s43586-021-00055-w
Box, G. E. P. (1976). Science and statistics. Journal of the American Statistical Association,
71 (356), 791–799. doi: 10.1080/01621459.1976.10480949
Box, G. E. P., & Draper, N. R. (1987). Empirical model-building and response surfaces. John Wiley
& Sons, Inc.
Brier, G. W. (1950). Verification of forecasts expressed in terms of probability. Monthly Weather
Review, 78 (1), 1–3.
Chen, Y., Li, X., Liu, J., & Ying, Z. (2018). Robust measurement via a fused latent and graphical
item response theory model. Psychometrika, 83 (3), 538–562. doi: 10.1007/s11336-018-9610-4
Consonni, G., Fouskakis, D., Liseo, B., & Ntzoufras, I. (2018). Prior distributions for objective
Bayesian analysis. Bayesian Analysis, 13 (2), 627–679. doi: 10.1214/18-BA1103
Contreras, A., Nieto, I., Valiente, C., Espinosa, R., & Vazquez, C. (2019). The study of psy-
chopathology from the network analysis perspective: A systematic review. Psychotherapy and
Psychosomatics, 88 (2), 71–83. doi: 10.1159/000497425
Dienes, Z. (2014). Using Bayes to get the most our of non-significant results. Frontiers in Psychology,
5 (781), 1–17. doi: 10.3389/fpsyg.2014.00781
Dobra, A., & Lenkoski, A. (2011). Copula Gaussian graphical models and their application to
modeling functional disability data. The Annals of Applied Statistics, 5 (2A), 969–993. doi:
10.1214/10-AOAS397
Eberhardt, F. (2017). Introduction to the foundations of causal discovery. International Journal of
Data Science and Analytics, 3 , 81-91. doi: 10.1007/s41060-016-0038-6
Epskamp, S., Borsboom, D., & Fried, E. I. (2018). Estimating psychological networks and their
accuracy: A tutorial paper. Behavior Research Methods, 50 , 195–212. doi: 10.3758/s13428
-017-0862-1
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 26

Epskamp, S., & Fried, E. I. (2018). A tutorial on regularized partial correlation networks. Psycho-
logical Methods, 23 (4), 617–634. doi: 10.1037/met0000167
Epskamp, S., Haslbeck, J. M. B., Isvoranu, A. M., & van Borkulo, C. D. (2022). Pairwise Markov
random fields. In A. M. Isvoranu, S. Epskamp, L. J. Waldorp, & D. Borsboom (Eds.), Network
psychometrics with R: A guide for behavioral and social scientists (pp. 93–110). Routledge,
Taylor & Francis Group.
Epskamp, S., Kruis, J., & Marsman, M. (2017). Estimating psychopathological networks: Be careful
what you wish for. PLoS One, 12 (e0179891). doi: 10.1371/journal.pone.0179891
Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recognition Letters, 27 (8), 861–874.
doi: 10.1016/j.patrec.2005.10.010
George, E. I., & McCulloch, R. E. (1993). Variable selection via Gibbs sampling. Journal of the
American Statistical Association, 88 (423), 881-889.
Giudici, P. (1995). Bayes factors for zero partial covariances. Journal of Statistical Planning and
Inference, 46 (2), 161–174. doi: 10.1016/0378-3758(94)00101-Z
Glymour, C., Zhang, K., & Spirtes, P. (2019). Review of causal discovery methods based on graphical
models. Frontiers in Genetics, 10 (524). doi: 10.3389/fgene.2019.00524
Gojković, V., Dostanić, J. S., & Ðurić, V. (2022). Structure of darkness: The dark triad, the ’dark’
empathy and the ’dark’ narcissism. Primenjena psihologija, 15 , 237 – 268.
Good, I. J. (1985). Weight of evidence: A brief survey. Bayesian Statistics, 2 , 249–270.
Guo, J., Levina, E., Michailidis, G., & Zhu, J. (2015). Graphical models for ordinal data. Biometrika,
24 (1), 183–204.
Haslbeck, J. M. B., & Waldorp, L. J. (2020). mgm: Estimating time-varying mixed graphical
models in high-dimensional data. Journal of Statistical Software, 93 (8), 1–46. doi: 10.18637/
jss.v093.i08
Hinne, M., Gronau, Q. F., van den Bergh, D., & Wagenmakers, E.-J. (2020). A conceptual intro-
duction to Bayesian model averaging. Advances in Methods and Practices in Psychological
Science, 3 (2), 200–215. doi: 10.1177/251524591989865
Hoeting, J., Madigan, D., Raftery, A., & Volinsky, C. (1999). Bayesian model averaging: A tutorial.
Statistical Science, 14 (4), 382–401.
Huth, K., de Ron, J., Goudriaan, A. E., Luigjes, K., Mohammadi, R., van Holst, R. J., . . . Marsman,
M. (2023). Bayesian Analysis of Cross-Sectional Networks: A Tutorial in R and JASP.
Advances in Methods and Practices in Psychological Science. doi: 10.1177/25152459231193
Huth, K., Keetelaar, S., Sekulovski, N., van den Bergh, D., & Marsman, M. (2023, 12). Simplifying
Bayesian analysis of graphical models for the social sciences with easybgm: A user-friendly
R-package. PsyArXiv. doi: 10.31234/osf.io/8f72p
Ioannidis, J. P. A. (2005). Why most published research findings are false. PLoS Medicine, 2 (e124).
doi: 10.1371/journal.pmed.0020124
Ising, E. (1925). Beitrag zur theorie des ferromagnetismus. Zeitschrift für Physik, 31 (1), 253–258.
doi: 10.1007/BF02980577
Jeffreys, H. (1939). Theory of Probability. Oxford, England: Clarendon Press.
Jeffreys, H. (1961). Theory of Probability (3rd ed.). Oxford, UK: Oxford University Press.
Jones, D. N., & Paulhus, D. L. (2014). Introducing the short dark triad (sd3) a brief measure of
dark personality traits. Assessment, 21 (1), 28–41. doi: 10.1177/1073191113514105
Jongerling, J., Epskamp, E., & Williams, D. R. (2022). Bayesian uncertainty estimation for Gaussian
graphical models and centrality indices. Multivariate Behavioral Research. doi: 10.1080/
00273171.2021.1978054
Kaplan, D. (2021). On the quantification of model uncertainty: A Bayesian perspective. Psychome-
trika, 86 (1), 215–238. doi: 10.1007/s11336-021-09754-5
Kass, R. E., & Raftery, A. E. (1995). Bayes factors. Journal of the American Statistical Association,
90 (430), 773–795. doi: 10.2307/2291091
Kass, R. E., & Wasserman, L. (1995). A reference Bayesian test for nested hypotheses and its
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 27

relation to the Schwarz criterion. Journal of the American Statistical Association, 90 (431),
928–934. doi: 10.1080/01621459.1995.10476592
Keysers, C., Gazzola, V., & Wagenmakers, E.-J. (2020). Using Bayesian factor hypothesis testing
in neuroscience to establish evidence of absence. Nature Neuroscience, 23 (7), 788–799. doi:
10.1038/s41593-020-0660-4
Kindermann, R., & Snell, J. L. (1980). Markov Random Fields and their Applications (Vol. 1).
Providence: American Mathematical Society.
Kruschke, J. K. (2011). Bayesian assessment of null values via parameter estimation and
model comparison. Perspectives on Psychological Science, 6 (3), 299–312. doi: 10.1177/
1745691611406925
Laumann, E. O., Marsden, P. V., & Prensky, D. (1989). The boundary specification problem in
network analysis. In L. C. Freeman, D. R. White, & A. K. Romney (Eds.), Research methods
in social network analysis. George Mason University Press.
Lauritzen, S. (2004). Graphical Models. Oxford: Oxford University Press.
Lenkoski, A. (2013). A direct sampler for G-Wishart variates. Stat, 2 (1), 119–128. doi: 10.1002/
sta4.23
Lindley, D. (2004). That wretched prior. Significance, 1 , 85–87. doi: 10.1111/j.1740-9713.2004.026
.x
Love, J., Selker, R., Marsman, M., Jamil, T., Dropmann, D., Verhagen, A. J., . . . Wagenmakers,
E.-J. (2019). JASP – graphical statistical software for common statistical designs. Journal of
Statistical Software, 88 (2), 1–17. doi: 10.18637/jss.v088.i02
Marsman, M. (2023). bgms: Bayesian variable selection for networks of binary and/or ordinal
variables [Computer software manual]. (R package version 0.1.0)
Marsman, M., Borsboom, D., Kruis, J., Epskamp, S., van Bork, R., Waldorp, L. J., . . . Maris,
G. K. J. (2018). An introduction to network psychometrics: Relating Ising network models to
item response theory models. Multivariate Behavioral Research, 53 (1), 15–35. doi: 10.1080/
00273171.2017.1379379
Marsman, M., & Haslbeck, J. M. B. (2023). Bayesian analysis of the ordinal Markov random field.
PsyArXiv. doi: 10.31234/osf.io/ukwrf
Marsman, M., Huth, K., Waldorp, L. J., & Ntzoufras, I. (2022). Objective Bayesian edge screening
and structure selection for Ising networks. Psychometrika, 87 (1), 47–82. doi: 10.1007/s11336
-022-09848-8
Marsman, M., Maris, G. K. J., Bechger, T. M., & Glas, C. A. W. (2015). Bayesian inference for
low-rank Ising networks. Scientific Reports, 5 (9050). doi: 10.1038/srep09050
Marsman, M., & Rhemtulla, M. (2022). Guest editors’ introduction to the special issue “net-
work psychometrics in action”: Methodological innovations inspired by empirical problems.
Psychometrika, 87 (1), 1–11. doi: 10.1007/s11336-022-09861-x
Mohammadi, A., & Wit, E. C. (2015). Bayesian structure learning in sparse Gaussian graphical
models. Bayesian Analysis, 10 (1), 109–138. doi: 10.1214/14-BA889
Mohammadi, R., & Wit, E. C. (2019). BDgraph: An R package for Bayesian structure learning in
graphical models. Journal of Statistical Software, 89 (3).
Morey, R. D., Hoekstra, R. H. A., Rouder, J. N., Lee, M. D., & Wagenmakers, E.-J. (2016). The
fallacy of placing confidence in confidence intervals. Psychonomic Bulletin & Review, 23 (1),
103–123.
Neal, Z. P., & Neal, J. W. (2023). Out of bounds? The boundary specification problem for centrality
in psychological networks. Psychological Methods, 28 (1), 179–188. doi: 10.1037/met0000426
O’Hagan, A. (2010). Kendall’s Advanced Theory of Statistic 2B. John Wiley & Sons.
Open Science Foundation. (2015). Estimating the reproducibility of psychological science. Science,
349 (6251). doi: 10.1126/science.aac4716
Pearl, J. (2009). Causality: Models, reasoning, and inference (Second ed.). Cambridge University
Press.
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 28

Robinaugh, D. J., Hoekstra, R. H. A., Toner, E. R., & Borsboom, D. (2020). The network approach
to psychopathology: A review of the literature 2008–2018 and an agenda for future research.
Psychological Medicine, 50 , 353–366. doi: 10.1017/S0033291719003404
Roverato, A. (2002). Hyper inverse Wishart distribution for non-decomposable graphs and its
application to Bayesian inference for Gaussian graphical models. Scandinavian Journal of
Statistics, 29 (3), 391–141. doi: 10.1111/1467-9469.00297
Rozanov, Y. A. (1982). Markov Random Fields. New York, NY: Springer-Verlag.
Ryan, O., Bringmann, L. F., & Schuurman, N. (2022). The challenge of generating causal hypotheses
using network models. Structural Equation Modeling: A Multidisciplinary Journal, 29 (6),
953–970. doi: 10.1080/10705511.2022.2056039
Sekulovski, N., Keetelaar, S., Haslbeck, J. M. B., & Marsman, M. (2023). Sensitivity analysis of prior
distributions in bayesian graphical modeling: Guiding informed prior choices for conditional
independence testing. PsyArXiv. doi: 10.31234/osf.io/6m7ca
Spirtes, P., Glymour, C., & Scheines, R. (2000). Causation, Prediction, and Search (Second ed.).
Cambridge, Massachusetts: MIT Press.
Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal
Statistical Society. Series B (Methodological), 58 (1), 267–288.
Vachon, D. D., & Lynam, D. R. (2016). Fixing the problem with empathy: Development and
validation of the affective and cognitive measure of empathy. Assessment, 23 (2), 135–149.
doi: 10.1177/1073191114567941
van Borkulo, C. D., Borsboom, D., Epskamp, S., Blanken, T. F., Boschloo, L., Schoevers, R. A., &
Waldorp, L. J. (2014). A new method for constructing networks from binary data. Scientific
Reports, 4 (5918). doi: 10.1038/srep05918
van de Schoot, R., Kaplan, D., Denissen, J., Asendorpf, J. B., Neyer, F. J., & van Aken, M. A. G.
(2014). A gentle introduction to Bayesian analysis: Applications to developmental research.
Child Development, 85 (3), 842–860. doi: 10.1111/cdev.12169
van Doorn, J., van den Bergh, D., Böhm, U., Dablander, F., Derks, K., Draws, T., . . . Wagenmak-
ers, E.-J. (2021). The JASP guidelines for conducting and reporting a Bayesian analysis.
Psychonomic Bulletin & Review, 28 (3), 813–826. doi: 10.3758/s13423-020-01798-5
Vanpaemel, W., & Lee, M. (2012). Using priors to formalize theory: Optimal attention and the
generalized context model. Psychonomic Bulletin & Review, 19 , 1047–1056. doi: 10.3758/
s13423-012-0300-4
van Ravenzwaaij, D., Cassey, P., & Brown, S. D. (2018). A simple introduction to Markov Chain
Monte–Carlo sampling. Psychonomic Bulletin & Review, 25 (1), 143–154. doi: 10.3758/
s13423-016-1015-8
Wagenmakers, E.-J. (2007). A practical solution to the pervasive problems of p values. Psychonomic
Bulletin & Review, 14 , 779–804. doi: 10.3758/BF03194105
Wagenmakers, E.-J., Lee, M. D., Rouder, J. N., & Morey, R. D. (2020). The principle of predictive
irrelevance or why intervals should not be used for model comparison featuring a point null
hypothesis. In C. W. Gruber (Ed.), The Theory of Statistics in Psychology – Applications,
Use and Misunderstandings. Cham: Springer.
Wagenmakers, E.-J., Love, J., Marsman, M., Jamil, T., Ly, A., Verhagen, J., . . . Morey, R. D. (2018).
Bayesian inference for psychology. Part II: Example applications with JASP. Psychonomic
Bulletin & Review, 25 (1), 58–76. doi: 10.3758/s13423-017-1323-7
Wagenmakers, E.-J., Marsman, M., Jamil, T., Ly, A., Verhagen, J., Love, J., . . . Morey, R. D. (2018).
Bayesian inference for psychology. Part I: Theoretical advantages and practical ramifications.
Psychonomic Bulleting & Review, 25 (1), 58–76. doi: 10.3758/s13423-017-1343-3
Wagenmakers, E.-J., Morey, R. D., & Lee, M. D. (2016). Bayesian benefits for the pragmatic
researcher. Current Directions in Psychological Science, 25 (3), 169–176. doi: 10.1177/
0963721416643289
Wagenmakers, E.-J., Wetzels, R., Borsboom, D., & van der Maas, H. L. J. (2011). Why psychologists
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 29

must change the way they analyze their data: The case of psi: Comment on Bem (2011).
Journal of Personality and Social Psychology, 100 (3), 426–432. doi: 10.1037/a0022790
Waldorp, L. J., & Marsman, M. (2022). Relations between networks, regression, partial correlations,
and the latent variable model. Multivariate Behavioral Research, 57 (6), 994–1006. doi: 10
.1080/00273171.2021.1938959
Williams, D. R. (2021). Bayesian estimation for Gaussian graphical models: Structure learning,
predictability, and network comparisons. Multivariate Behavioral Research, 56 (2), 336–352.
doi: 10.1080/00273171.2021.1894412
Williams, D. R., & Mulder, J. (2020a). Bayesian hypothesis testing for Gaussian graphical mod-
els: Conditional independence and order constraints. Journal of Mathematical Psychology,
99 (102441).
Williams, D. R., & Mulder, J. (2020b). BGGM: Bayesian Gaussian graphical models in R. Journal
of Open Source Software, 5 (51), 2111. doi: 10.21105/joss.02111
Williams, D. R., Rhemtulla, M., Wysocki, A. C., & Rast, P. (2019). On nonregularized estimation
of psychological networks. Multivariate Behavioral Research, 54 (5), 719–750. doi: 10.1080/
00273171.2019.1575716
BAYESIAN TESTS FOR CONDITIONAL INDEPENDENCE 30

Appendix
Prior Distributions for MRF Models implemented in R packages
The Bayesian analysis of an MRF model requires specifying two sets of prior distri-
butions.

1. Priors on the structure

The prior probabilities on the network structure p(Ss ) can be expressed by specifying
a prior probability on the possible value for each binary indicator variable γij . This is
achieved by assuming that each edge follows an independent Bernoulli distribution with a
prior inclusion probability πij . The R packages bgms (Marsman, 2023) for analyzing MRF
for binary and ordinal data and BDgraph (R. Mohammadi & Wit, 2019) for analyzing GGMs
both provide this as the default option for the prior on the network structure. Setting this
prior with πij = 0.5 for all edges is considered an uninformative or objective choice, and this
option is also referred to as the uniform prior on the structure (e.g., Marsman et al., 2022).
Of course, there are other prior options for the network structure that take into account
the number of present edges (i.e, the complexity of the structure); we refer the interested
reader to Huth, de Ron, et al. (2023) for an accessible introduction to these priors and
to Sekulovski, Keetelaar, Haslbeck, and Marsman (2023) for a detailed discussion of prior
selection with particular emphasis on the priors implemented in the R package bgms.

2. Priors on the edge weights

We also need to specify priors on the edge weight parameters in Θ. The R package
BDgraph specifies a g-Wishart distribution (Roverato, 2002) on the precision matrix (i.e., the
inverse of the covariance matrix) containing the (untransformed) edge weight parameters
for the GGM. The g-Wishart distribution takes two parameters (i) the degrees of freedom d,
which is by default set to d = 3, and (ii) a scale matrix D, set by default to an uninformative
p × p identity matrix. The R package BGGM (Williams & Mulder, 2020b), which can also
be used to analyze GGMs, specifies either a Wishart or a Matrix F prior on the precision
matrix (Williams & Mulder, 2020a). The (g-)Wishart priors are conjugate to the precision
matrix and assure a posterior density function on the space of positive semi definite matrices.
Note that the R package BGGM does not stipulate priors on the network structure since it
assumes that all edges are present a priori. Finally, the R package bgms specifies prior
distributions on the individual edge weights given the value of the edge indicator variable
γij , i.e., p(θij | γij ). In other words, if γij = 0, the edge weight is set to zero, and if γij = 1,
the edge is given a specific (diffuse) prior distribution (e.g., a Cauchy distribution). For
more details, see Marsman and Haslbeck (2023) and Sekulovski et al. (2023).