Questions tagged [controlling-for-a-variable]

A control variable is one which is included in a model primarily for its impact on the total model rather than for its own substantive interest.

Filter by
Sorted by
Tagged with
1 vote
1 answer
22 views

randomisation issue/crisis

I have randomized into three groups using a randomization aid and groups were examined at 2 time points. however in the analysis found that there was a significant difference at baseline in one of the ...
2 votes
1 answer
22 views

Can covariates be included in multiple regression?

I am interested in understanding if two predictors variables are associated with treatment outcome (DV). I included two covariates that 1) differed across my levels of IV (age) and 2) another that is ...
2 votes
1 answer
1k views

Adding Controls to Staggered Difference-in-Difference Regressions in Stata

I am running a staggered diff-in-diff model, looking at legalization's effect on various variables. For context, only a percentage of all states have legalized, and the year they legalized differs ...
3 votes
1 answer
234 views

Multicollinearity and control variables dilemma

I had some superficial understanding of multicollinearity, that two highly correlated variables in the regression model are not what we want, as the estimated coefficient would be biased. Control ...
1 vote
1 answer
53 views

Multilevel Modeling and Alternatives for Repeated Measures Design

I am conducting a study on the effect of different characteristics of urban space on human physiological response. The study is a "repeated measures" design. For this example, each of 5 ...
0 votes
0 answers
11 views

Question regarding control variables in a serial mediation model

i encountered questions regarding running a serial mediation model in mplus. my IV is ERS(excessive reasssurance seeking) DV is depression (DEP), my proposed two mediators are INU (internet use) and ...
1 vote
0 answers
36 views

How do I regress income quartiles against each other?

I'm looking to find out whether an attitude differs across income quartiles. My supervisor has mentioned dummy coding and regressing the quartiles against each other, however, I'm sort of at a loss as ...
0 votes
1 answer
71 views

ANOVA/ANCOVA results in SPSS

I am unsure about how to interpret the output box I get doing an ANOVA in SPSS. That's the "Tests of between-subject effects" box in particular. I had 2 groups (A and B), and 2 results (...
2 votes
1 answer
6k views

Do control variables need to have a correlation to both the independent variable and the dependent variable?

I have been wondering about this: Do control variables need to have a correlation to both the independent variable and the dependent variable? E.g. I want to check the effect of Education (...
5 votes
2 answers
196 views

Is there a technique similar to multiple regression which does not require linearity?

Background: We have biological gene expression data of cells (from a single-cell experiment). So, the data is in the form of a gene by cell matrix, where each value is the expression of a specific ...
5 votes
0 answers
92 views

Is it meaningful to control for variables that are correlated with the independent and dependent variables?

Statistical training from the National Institute of Child Health and Human Development states: A general (usually unquestioned) assumption is that differences on "nuisance" variables that ...
2 votes
1 answer
257 views

Can we isolate the effect of two different control variables in a semi-partial correlation?

I know that we are able to use a partial correlation when we want to correlate X and Y but Z affects both of them and that we may use a semi-partial correlation when we want to correlate X and Y and ...
1 vote
1 answer
276 views

How to check for relevant covariates in difference-in-difference regression?

I am conducting a diff-in-diff regression over two periods of time with the treatment in between. Next to the dummy variable time ('Post'), I have 3 groups for the other variable and therefore 2 dummy ...
1 vote
1 answer
38 views

Using interaction terms in causal models

I have the following model: LOS (length of stay) = b0 + b1 Age + b2 Gender + b3 BMI where Age and Gender are control variables, and my main IV is BMI. And this model isn't made for predictive purpose; ...
4 votes
1 answer
60 views

What is an intuitive explanation for a lay-person why controlling for a collider is bad practice while controlling for a confounder is good practice?

Cinelli et al., 2022 and Wysnocki et al., 2022 describe in technical terms how controlling for a collider can lead to biased estimates. If one needs to explain why one should not control for a ...
1 vote
1 answer
120 views

What is an “independent” effect?

According to one study by Wang and colleagues (2013), smoking cigars has “a positive and independent” effect on testosterone. What is an independent effect and what is the significance of one?
4 votes
1 answer
85 views

When is "controlling for" effects nonsensical?

I am not as knowledgeable about statistics as I would like and I am looking for resources to help me understand what kinds of questions are meaningful and which are nonsensical. We are developing some ...
1 vote
0 answers
20 views

Serial multiple mediated model in lavaan not calculating standard errors, Z-values, and P values for defined parameters

I am trying to run the following model in Lavaan, in which AM, RM, and CU are latent variables. It also has three control variables (CV1, CV2, and CV3). I have written the code that is included at ...
2 votes
1 answer
58 views

Controlling for a variable when estimating interaction effects

Consider the relationships among four variables: sex, activity, height, and ...
1 vote
1 answer
1k views

How do you control for a variable in SEM with mediation? What do I regress onto my control variables?

I have an SEM model that tests for mediation. There is 1 independent variable, 1 mediator, and 2 dependent variables. All these variables are continuous. I want to control for the following variables:...
0 votes
0 answers
16 views

Multiple proportion test

I want to test which week day is more important for a specific client, given a historical communication in different days and with open email or not . for example: monday Tuesday Wednesday ... email ...
2 votes
2 answers
294 views

What analysis do I do? (Wage, education, gender)

I'm doing research on pay in function of degree (master vs. doctorate) and gender (man vs. woman). My research consists of 1 dependent variable: salary (salary varies from 210 euro to 12.000 euro in ...
4 votes
2 answers
209 views

According to DAG theory, why controlling for this variable doesn't close the backdoor path opened by controlling for the collider?

I have made the following model in DAGitty: Where $X_2$ is controlled for. DAGitty says: The total effect cannot be estimated due to adjustment for an intermediate or a descendant of an intermediate....
1 vote
0 answers
12 views

Including an interaction variable which doesn't vary within a key control variable

Let’s say we want to check the effect of the interaction of education level and gender of the salesperson on number of units sold by them. We have thousands of sales people in the data, but they only ...
0 votes
0 answers
15 views

Exogeneity Check of Control Variables? Covariance

I have performed an exogeneity check on my control variables to assess their validity before incorporating them into my regression analysis. Upon conducting the analysis, I observed a significant ...
3 votes
1 answer
99 views

Is it a bad idea to use a variable that is strongly correlated with my independant variable of interest as a control variable?

I am currently looking into the correlation between academic freedom (my independant variable) and university rankings (my dependant variable) using OLS. I find a negative significant correlation, but ...
0 votes
1 answer
58 views

Logic of including an interaction term as a control variable?

I am reading a paper and it has the following simple model: Y = a + b1 + b2 + b1*b2 + e. The author seems to be interested in b1, not the interaction term. How do we interpret b1 in this case? I'm a ...
1 vote
0 answers
42 views

How to account for data's margins of error in a regression?

Apologies if this is too simple of a question. I work with population data from the census, which provides a margin of error for all their variables. Reviewers have asked that I account for the MOE in ...
0 votes
0 answers
10 views

Control variables in interaction effects [duplicate]

I'm new to econometrics and this might be a dumb question.. but if I were to include an interaction term, do I also have to input the variables separately as a control variable? For example, Y = a + ...
0 votes
0 answers
32 views

What makes a good control variable?

I am currently studying whether academic freedom (independant variable) has an impact on university rankings (independant variable). So far, my model is only composed of these two variables, as well ...
0 votes
0 answers
68 views

How does R know which variable is the main variable of interest?

I'm starting to get a little confused about how R knows which variable is the main variable and which others are simply controls. For example, let's say there are three variables Y, X1, and X2, and I'...
0 votes
0 answers
13 views

Influence of demographics/background on part worth utilities in conjoint analysis

I’m currently working on a conjoint analysis (CBC) about e-commerce fulfillment. The idea is to find out (among other things) whether customer characteristics (gender, age, home-presence, …) influence ...
1 vote
1 answer
49 views

do not need many controls with big data?

I am looking at a paper which uses a large panel data, 1 million observations, a dozen variables. I recall that in a discussion another one has the following comments: In structural models like ...
2 votes
1 answer
33 views

Controlling parents education that correlates with each other

I want to include both parent's education variables as control variables in my estimation about the effects of maternal bargaining power to child's educational attainments. They are stated as ...
1 vote
0 answers
23 views

Does it make sense to "regress out on one variable" during permutation testing?

In our lab we have developed a non parametric permutation test which assesses whether a feature $f:G\to \mathbb{R}$ respects a given symmetric binary relation $A\subset G\times G$. Regard $A$ as being ...
2 votes
0 answers
227 views

Why add control variables in the second stage of an IV regression?

I am reading an article on the effect of income on energy choice (e.g., choice between traditional sources such as charcoal and modern ones such as electricity). The authors instrument income with ...
0 votes
1 answer
204 views

Difference between controlling for other variables in additive models vs. interaction model

In a regression model with multiple predictors, the word "control" is used to refer to the inclusion of other variables than the one relevant to a specific question (e.g., the effect of ...
0 votes
1 answer
119 views

Controlling for individual differences in a repeated-measures design?

I have been analysing a study where around 500 participants each were given a survey, so there were repeated measurements of each participant. Participants were male and female uni students who were ...
-1 votes
1 answer
40 views

Dependent variable "controlled" by a binary variable?

in my research the DV should be measured by means of an intent scale (Individual Entrepreneurial Intent Scale). For "controlling" if the answer in DV is correct, I want to include a binary ...
7 votes
2 answers
829 views

Can age ever be confounded if it is the independent variable in an observational study?

This is a follow-up to my previous question: "Basic understanding of control variables in observational studies" I have understood the answer, but am still trying to figure out what I think ...
1 vote
1 answer
47 views

Controlling for an effect by adding it as covariate in R

I have a dataset like this: individual1 individual2 Bray_Curtis Dyad_type proximity matriline id1 id2 Aapi App 0.47 1 0.14 1 1 2 Aapi Eis 0.60 2 0.03 0 1 4 Potj Popp 0.50 1 0.11 1 3 5 Aapi Potj 0....
7 votes
2 answers
556 views

Can we just "pre-test" the backdoor criterion?

I am trying to use DAGs to think more carefully about my regression models. I have a question about the "backdoor criterion", as usually seen in the DAG below (we are interested in the ...
1 vote
1 answer
130 views

MANOVA (SPSS) test - determine control variables in moderation analysis

I am examining the moderating effect of IV1 in the relationship between IV2 and DV using hierarchical regression analysis. The study has 2 demographic variables: gender and age. To determine if it is ...
3 votes
1 answer
3k views

Ceteris paribus relationship and parameters in linear regression models

I'm reading Wooldridge’s Introductory Econometrics text and I'm confused about this passage here: "The most difficult issue to address is whether model (y = b0 + b1x + u) really allows us to draw ...
3 votes
1 answer
872 views

Controlling for a variable: t-test or covariate?

Lets say I have some data that looks like this: ...
2 votes
0 answers
28 views

How to study the impact of time in a mixed-factorial design?

I ran a study to compare 2 different computer interfaces (say A and B) meant to be used in an unfamiliar environment. I want to compare: How much effort participants took to complete a certain task ...
5 votes
2 answers
682 views

Basic understanding of control variables in observational studies

I'd like to understand the basic logic for control variables in observational studies. According to Wikipedia, one should control for confounding variables, which "in this context means other ...
1 vote
1 answer
186 views

Data points for some control variables missing in regression - still feasible?

I'm currently working on an event study to examine abnormal returns. In the first step, I've calculated abnormal returns in regards to a certain type of company event, consisting of roughly 13,000 ...
1 vote
0 answers
61 views

Is there a formal methodology for controlling a variable in a tree based model?

I am currently building a xgboost model to classify data into 6 categories of risk for insurance policies. I have 5 years worth of policy holder data, including policy year. When building a GLM for a ...
1 vote
0 answers
498 views

Are control variables always exogenous?

My professor said that control variables are generally thought to be exogenous in an OLS context, i.e. uncorrelated with the main independent variable in my regression. However, I think that they are ...