Multiple Indicators Multiple Causes Model Research Articles

Information technologies and product-service innovation: The moderating role of service R&D team structure

Drawing on the ambidexterity and organizational design theoretical lenses this article analyzes the interplay between R&D team structure, firm’s Information Technology (IT) processes deployment and innovation outcomes. The evidence presented herein upholds the importance of IT and R&D team structure for strategic decisions and to better exploiting firm’s innovation capabilities. Concretely, we argue that R&D team structure (centralized vs formalized vs autonomous) moderates the relationship between IT processes and innovation because it influences the way in which IT is utilized. Considering these facts, we focus on a specific type of innovation, Product-Service Innovation (PSI), largely underexplored despite being increasingly important in modern manufacturing companies. PSI differs from other technological innovations in that it involves continuous engagement with customers and logistics. Through estimation of a Multiple-Indicators Multiple-Causes (MIMIC) model with a unique sample of 352 Manufacturing Multinational Enterprises (MMNEs), we find that customer and logistics IT processes are positively linked to higher levels of PSI and that, as hypothesized, service R&D team structure moderates this relationship. In firms with autonomous R&D teams, customer-based IT processes lead to higher PSI levels, whereas in firms with formalized R&D teams, logistics-based IT processes is conducive to higher PSI levels. IT processes are not an input of PSI in centralized service R&D teams.

Psychometric properties of the Inventory of Study Habits CASM-85: a multicenter study with Peruvian high school students

Objective: To assess the internal structure, measurement invariance, criterion validity and reliability of the CASM-85 in Peruvian high school students. Materials and methods: Two thousand seventy-five (2075) students of both sexes, aged 11 to 18 years, from public and private high schools of Arequipa, Chimbote, Lima and Pucallpa were evaluated. An item-item and item-rest correlation analysis for each dimension of the CASM- 85, a confirmatory factorial analysis and MIMIC (Multiple Indicator, Multiple Cause) models were used to evaluate the measurement invariance across the groups according to their sex, age, school and city. Reliability was estimated using alpha and omega coefficients. The criterion validity between the CASM-85 and academic performance was determined using Pearson’s r and regression models. Results: The heterogeneous items were removed, obtaining a 45-item short form compared to the original 53-item form. In the original form, CFI and TLI were < .9 in all the cities, whereas in the short form, CFI and TLI were < .9 only in Chimbote. SRMR and RMSEA were ≤ .10 and ≤ .09 in the original and short forms, respectively. The MIMIC models confirmed the measurement invariance in all variables, with the exception of the original form between cities (|ΔCFI| = .09, |ΔTLI| = .08). Reliability was moderate for both the original (Ω ≥ .68; α ≥ .81) and short forms (Ω ≥ .77; α ≥ .84). Pearson’s r showed a positive and weak correlation in the original (r = .17, p < .001) and short (r = .18, p < .001) forms. Conclusion: The CASM-85 short form (45 items) has shown better psychometric properties than its original full form.

Evaluation of E-business micro and macro determinants by Multiple Indicators Multiple Causes model

The methods employed for the research include comparative and systematic literature analysis, statistical data analysis, expert evaluation and the Multiple Indicators Multiple Causes (MIMIC) model. There are several research gaps in the field of e-business: 1) there is no simplified factor model which would denote the development of e-business, including its causes and consequences, and which, due to its universality, could be adapted to any country; 2) the studies of e-business development lack a comprehensive assessment which would consider this phenomenon in both quantitative and qualitative terms.&#x0D; On the basis of the theoretical analysis, a model for assessment of e-business development has been formed. The results of the empirical evaluation have revealed that in the group of micro environmental determinants development of e-business in Lithuania is, to the largest extent, determined by business managers’ motivation to reach the defined aims and their positive viewpoint that e-business may become a perfect auxiliary tool to ensure business competitiveness. In the group of macro environmental determinants, the determinants of socio-cultural environment can be indicated as the major ones that affect the development of e-business. Application of the MIMIC model has enabled to identify the following macro economical determinants (for the euro area): real effective exchange rate for the euro area; level of households with the Internet access; quantity of individuals using the Internet for ordering goods or services; government deficit/surplus; youth employment, and individuals using the Internet for interaction with public authorities.

Private Forest Owners’ Social Economic Profiles Weakly Influence Forest Management Conceptualizations

Although several private forest owner studies have dealt with how private forest owners understand forest management, little is known about the determinants of specific forest management concepts. The study expands previous latent variable models of the perception of forest management by European private forest owners by looking at how age, income, education, annual cut, and holding size and type influence specific understandings of forest management. We applied a multiple indicators multiple causes (MIMIC) structural equation model on a representative sample of 754 private forest owners from Slovenia. The MIMIC model confirmed the influence of six covariates on three concepts of forest management: the maintenance concept, the ecosystem-centered concept, and the economics-centered concept. The strongest determinants of perception were education and holding type. The maintenance concept was predominantly associated with less educated older full-time or part-time farmers working on smaller family farms and doing regular cuts. The perception of forest management as an economics-centered activity increased with increased education and dependence on income from intensive cuts. The ecosystem-centered concept was most strongly associated with younger, better-educated owners with smaller holdings and, surprisingly, not to non-farmers but to small-scale family farmers. However, the proportion of the variance of latent variables explained by the six covariates was low, ranging from 2.4% to 5.1%. Taking into account the influence of education and holding type on private forest owners’ perception of forest management, by increasing the level of education and raising the proportion of absentee owners in Europe, we expect a shift from the maintenance concept toward either an economics-centered or ecosystem-oriented concept for forest management. Despite the weak influence of private forest owners’ social economic profiles on forest management conceptualizations, governments should be aware of the trend and actively seek to prevent the polarization of forest management concepts.

The relationship between personality traits and psychosomatic complaints in a sample of Iranian adults

BackgroundIt has been suggested that personality traits may be related to the psychological conditions of people. The current study aimed to explore the relationship between personality traits and psychosomatic complaints. MethodsThis study was carried out on a sample of Iranian adults living in Isfahan province. The NEO Five- Factor Inventory questionnaire was used to assess personality traits. A 30-item comprehensive questionnaire was used to assess somatoform symptoms and factor analysis was performed to construct psychosomatic complaints profiles. Data was analyzed by Multiple Indicators Multiple Causes (MIMIC) model, using R software. ResultsIn total, 4763 participants were recruited 44.2% males and 55.8% females, with a mean age of 36.54 ± 8.04 years. The results of the MIMIC model showed that there is a significant association between the dimensions of personality (neuroticism, extroversion, openness, agreeableness and conscientiousness) and dimensions of psychosomatic complaints profiles (gastrointestinal, neuro-skeletal, pharyngeal-respiratory and psychological) (P < 0.05). In addition, we found that characteristics such as age, sex, marital status, education, physical activity, and body mass index are related to psychosomatic complaints (P < 0.05) and can affect the association between personality traits and psychosomatic complaints. ConclusionsGiven the relationship between personality traits and psychosomatic complaints, it is possible to provide preventive and therapeutic interventions for psychosomatic complaints based on personality traits specific to each individual.

Early childhood predictors of global competence in adolescence for youth with typical development or intellectual disability

Background and aimsWe aimed to determine whether a second-order global competence latent factor could be identified as underlying relations between adolescent mental health, social skills, and academic functioning. A secondary aim was to test whether early childhood characteristics predict adolescent global competence. A final aim was to test differences in these models across youth with typical cognitive development (TD) or intellectual disability (ID). Methods and proceduresParticipants were 246 youth with TD (n = 148) or ID (n = 98), with assessments from early childhood (3, 4, 5 years) and adolescence (13, 15). These youths' parents and teachers provided measures. A Multiple Indicator, Multiple Causes (MIMIC) model was tested using structural equation modeling, in which parenting, maternal depression, and emotional dysregulation in early childhood were entered as predictors of adolescent global competence. Outcomes and resultsA second-order global competence factor emerged, and was predicted by early childhood variables. The final MIMIC model demonstrated excellent fit. Negative parenting in early childhood predicted lower adolescent global competence for both TD and ID youth. Maternal depression predicted adolescent global competence only for youth with ID, while emotion dysregulation predicted only for youth with TD. Conclusions and implicationsResults have implications for longitudinal mechanisms of influence and early intervention targets for specific populations.

Are individual differences quantitative or qualitative? An integrated behavioral and fMRI MIMIC approach

In cognitive neuroscience there is a growing interest in individual differences. We propose the Multiple Indicators Multiple Causes (MIMIC) model of combined behavioral and fMRI data to determine whether such differences are quantitative or qualitative in nature. A simulation study revealed the MIMIC model to have adequate power for this goal, and parameter recovery to be satisfactory. The MIMIC model was illustrated with a re-analysis of Van Duijvenvoorde et al. (2016) and Blankenstein et al. (2018) decision making data. This showed individual differences in Van Duijvenvoorde et al. (2016) to originate in qualitative differences in decision strategies. Parameters indicated some individuals to use an expected value decision strategy, while others used a loss minimizing strategy, distinguished by individual differences in vmPFC activity. Individual differences in Blankenstein et al. (2018) were explained by quantitative differences in risk aversion. Parameters showed that more risk averse individuals preferred safe over risky choices, as predicted by heightened vmPFC activity. We advocate using the MIMIC model to empirically determine, rather than assume, the nature of individual differences in combined behavioral and fMRI datasets.

The Colorado Cognitive Assessment (CoCA): Development of an Advanced Test of Cognitive Status

Abstract Objective Current brief tests of cognition provide a rough indication of overall cognitive functioning that assist with making gross clinical judgments (e.g., demented vs. not demented). The purpose of the present study was to develop and preliminarily validate a brief tool, The Colorado Cognitive Assessment (CoCA) to facilitate early and accurate diagnosis of mild and atypical presentations of dementia. A related goal was to compare its psychometric properties with The Montreal Cognitive Assessment (MoCA). Method Participants were 150 community dwelling adults over the age of 50 without a known mental health or neurological condition. Confirmatory factor analysis (CFA) was used to assess model fit of the CoCA and MoCA. Measurement invariance (MI) was evaluated using the multiple-indicators multiple-causes modeling (MIMIC) approach. Results The CFA model of the CoCA revealed excellent fit; χ2(44) = 47.506, p = .332; CFI = 0.987; TLI = 0.983; RMSEA = 0.023 (90% CI [0.000, 0.061]); SRMR = 0.048. MI analyses revealed that items on the CoCA were invariant to sex, age, education, and mood. In comparison, a CFA model of the MoCA had worse fit; χ2(14) = 28.536, p = .012; CFI = 0.817, TLI = 0.725; RMSEA = 0.083 (90% CI [0.038, 0.127]); SRMR = 0.062; and was biased by age, education, and depressive symptomatology. The global factor score reliability of the CoCA (r = .84) was better than the MoCA (r = .74). Conclusion Results provide preliminary evidence for the CoCA as a reliable and comprehensive cognitive instrument with further cross-sectional and longitudinal research needed for its validation.

Serum Adiponectin is Related to Dementia.

The adipokine adiponectin (APN)'s role in Alzheimer's disease (AD) is controversial. Some studies suggest APN is neuroprotective while others propose it has harmful effects. We have used Multiple Indicators Multiple Causes (MIMIC) models to evaluate the effects of serum protein biomarkers on cognitive performance in the Texas Alzheimer's Research and Care Consortium (TARCC) (Royall DR, Bishnoi RJ, Palmer RF. Serum IGF-BP2 strongly moderates age's effect on cognition: a MIMIC analysis. Neurobiol Aging. 2015;36:2232-2240; Bishnoi RJ, Palmer RF, Royall DR. Vitamin D binding protein as a serum biomarker of Alzheimer's disease. J Alzheimers Dis. 2015;43:37-45; Bishnoi RJ, Palmer RF, Royall DR. Serum interleukin (IL)-15 as a biomarker of Alzheimer's disease. PLoS One. 2015;10:e0117282). MIMIC models were constructed and replicated in randomly selected 50% splits of TARCC's data (Group 1 N = 1,691; Group 2 N = 1,690) and used to evaluate the relationship between serum APN levels and cognition. Our approach has been to divide general intelligence (Spearman's g) (Spearman C. The Abilities of Man: Their Nature and Measurement. 1932) into two latent variables, δ (ie, a dementia-specific phenotype representing the disabling fraction of cognitive variance) and g prime (g') (ie, the residual non-disabling fraction). Only effects on δ are likely to be dementing. Serum APN was significantly related to δ scores (r = .10, p = .015). APN had no significant effect on g' (r = -.25, p = .66), nor did it have any independent direct effects on cognitive performance. These results were replicated across random subsets (ΔCHISQ = 2.8(7), p > .90). APN's effect on cognition is mediated through intelligence (ie, δ), likely to be disabling, and therefore to mediate one or more dementing processes. We have previously shown APN to partially mediate age's-specific effect on δ (Royall DR, Al-Rubaye S, Bishnoi R, Palmer RF. Serum protein mediators of dementia and aging proper. Aging (Albany NY). 2016;8:3241-3254). However, because the current model is age adjusted, APN must mediate one or more additional age-independent dementing process(es), possibly AD.

Do banking sector and stock market development matter for economic growth?

Financial development as a concept is multifaceted with no clear measurement or definition. Inference via individual proxies may result in an incomplete understanding of the relationship between financial development and economic growth, since sole proxies are unlikely to capture the true capacity of financial development. To address this issue, this paper utilizes a multiple indicators multiple causes (MIMIC) model to create a more complete measure of financial development. In doing this, we treat banking sector and stock market developments as two latent indicators of financial development and use the MIMIC model to predict them which are used as their proxies. Using data from 101 countries over the period 1990–2014, we use the predicted values of the two latent variables as regressors, among other controls, in the growth regression. We find a robust negative relationship between banking sector development and economic growth, whereas the effect of stock market development on economic growth is positive up to a threshold after which the effect becomes negative.

Measurement Invariance of the UPPS-P Impulsive Behavior Scale Across Age and Sex Across the Adult Life Span.

Impulsive personality is a widely used construct for the prediction of multiple clinical problems; however, research has often disregarded important differences in its conceptualization and measurement across age and sex. The goals of this article are to test the invariance of, to compare mean differences in, and to examine the differential validity of the Urgency (negative), Premeditation (lack of), Perseverance (lack of), Sensation Seeking, with added subscale of Positive Urgency (UPPS-P) Impulsive Behavior Scale across adult age and sex. The data for this study were obtained from 799 participants in the Nathan Kline Institute's Rockland Sample. Multigroup invariance analysis indicated full invariance of the UPPS-P Impulsive Behavior Scale across age and sex. However, invariance analysis as a function of age using Multiple-Indicators Multiple-Causes modeling identified three items with differential item functioning. Sensation Seeking was higher in males and lower as age increased. In general, age and sex did not moderate the relationships between UPPS-P traits and risk-taking behaviors, with the exception of cannabis use. Overall, the results of this study support that the UPPS-P scale can be validly used to assess and compare impulsive personality across the adult life span and sex. We suggest the removal of Items 5, 28, and 51 when using the UPPS-P with older adults to be overly cautious to age invariance effects.

Ograničavajući čimbenici izostanka aktivnosti na šumoposjedima u Republici Hrvatskoj – MIMIC model

The purpose of this paper was to determine the share of active and inactive forest owners in Croatia and to identify potential causes for underuse of wood resources and their most important predictors based on inactive forest owners sample. We interviewed 1007 randomly selected forest owners on Croatian territory with a telephone survey. Of the total sample, 60% of forest owners believe that they efficiently manage their forests while 40% are inactive (Table 1). In order to determine the most important reasons and causes for inactive forest owners not cutting more, additional questions were asked. They were provided by 19 statements describing possible reasons for underuse of wood resources. Then, on Likert’s scale from 1 (not important at all) to 5 (very important) they expressed their level of agreement with a particular statement (Table 2). The statements were condensed into three dimensions by explanatory factor analysis with eigenvalue approach. Factors were interpreted as: (1) factor of qualifications and knowledge constraints, (2) factor of passiveness and forest conservation and (3) factor of physical constraints (Table 2). Statistically acceptable model was established by confirmatory factor analysis approach (Figure 2). The pronounced processes of structural changes in forest ownership composition, as well as in socio-demographic and forest property characteristics, are evident across Europe consequently influencing the intensity of private forest management. Within the aforementioned, we used gender, age, education, working status, the distance between the forest and the place of residence, the size of the settlement and the knowledge of forest property boundary lines as independent variables for management constraining factors model construction (Figure 1). Socio-demographic and forest property characteristics have been modeled with the extracted management constraining factors by applying a Multiple Indicators Multiple Causes (MIMIC) model (Figure 3). MIMIC model revealed that gender, age, and lack of boundary lines knowledge were the most important predictors for management constraining factors (Table 4). The notable part of forest owners in Croatia do not manage their forests because of insufficient knowledge and qualifications.

MIMIC approach to assessing differential item functioning with control of extreme response style

Likert or rating scales may elicit an extreme response style (ERS), which means that responses to scales do not reflect the ability that is meant to be measured. Research has shown that the presence of ERS could lead to biased scores and thus influence the accuracy of differential item functioning (DIF) detection. In this study, a new method under the multiple-indicators multiple-causes (MIMIC) framework is proposed as a means to eliminate the impact of ERS in DIF detection. The findings from a series of simulations showed that a difference in ERS between groups caused inflated false-positive rates and deflated true-positive rates in DIF detection when ERS was not taken into account. The modified MIMIC model, as compared to conventional MIMIC, logistic discriminant function analysis, ordinal logistic regression, and their extensions, could control false-positive rates across situations and yielded trustworthy true-positive rates. An empirical example from a study of Chinese marital resilience was analyzed to demonstrate the proposed model.

The Non‐Observed Economy in Portugal: The monetary model and the MIMIC model

AbstractThis paper offers estimations for the Portuguese path of the Non‐Observed Economy (NOE), in the period 1970–2015, through two seminal approaches: monetary method and the Multiple Indicators Multiple Causes (MIMIC) model. It is observed that the tax burden and social benefits are its main causes. Then, to get a more in‐depth understanding of the phenomenon, it provides a study of the Granger causality between the NOE and the official Gross Domestic Product (GDP), emphasizing the implications of the NOE on the Portuguese economy. Evidence has been found for the existence of bidirectional causality between the NOE and the GDP, suggesting that the formal economy affects the NOE, and conversely that the NOE affects the economic growth.

Condition evaluation of urban metro shield tunnels in Shanghai through multiple indicators multiple causes model combined with multiple regression method

Many existing condition evaluations of metro shield tunnels are developed based on the experts’ judgments or single deformation/defect measurements. In addition, the influence of factors such as buried depth and age on the condition of shield tunnels was not quantified in previous research. This study proposed a comprehensive condition index, the Tunnel Serviceability Index (TSI), through the Multiple Indicators Multiple Causes (MIMIC) model combined with a multiple regression method. The contributions of the deformation/defect variables and the influencing factors to the TSI were evaluated. The observed deformation/defect variables included average relative settlement, differential settlement, average convergence ratio, water leakage area, cracking length, and spalling area. The influencing factors included the operation age, burial depth, and so on. The relationship among the TSI, deformation/defect variables, and influence factors was quantified, and an expression for the TSI was obtained. The paired-sample t test demonstrated that there were no significant differences between the experts’ ratings in our group’s previous research and the tunnel condition evaluation used by the MIMIC model in this study at the significance level α = 0.05. Among the six observed deformation and defect variables, approximately 95% of the TSI deterioration was caused by the increase of convergence ratio, relative settlement and differential settlement. Finally, the case of Shanghai Metro Line No. 1 was investigated to evaluate its condition.

Assessing the measurement invariance of the RCADS-25 questionnaire across gender and child-parent dyads in the presence of multilevel data.

This study aims to assess the measurement invariance of the short version of the Revised Child Anxiety and Depression Scale (RCADS-25) across male and female adolescents along with child self-reports and parent proxy-reports. The sample consisted of 2039 adolescents and their parents from 51 public and private schools in Shiraz (southern Iran). The measurement invariance of the RCADS-25 was investigated by the multilevel multiple-indicators multiple-causes model. In addition, the reliability of the questionnaire was examined by Cronbach's alpha coefficient, and its validity was evaluated by exploratory and categorical confirmatory factor analysis. Although the RCADS-25 includes five (20%) non-invariant items across male and female adolescents, their impact is negligible at the scale level. In addition, 10 out of 25 items (40%) were non-invariant across self- and proxy-reports, which did not have substantial effect at the scale level. Based on our findings, the instrument can be used, albeit with caution, for comparing depression and anxiety scores between male and female adolescents as well as child self-reports and parent proxy-reports.

An IRT-Multiple Indicators Multiple Causes (MIMIC) Approach as a Method of Examining Item Response Latency.

The analysis of response time has received increasing attention during the last decades, since evidence from several studies supported the argument that there is a direct relationship between item response time and test performance. The aim of this study was to investigate whether item response latency affects person's ability parameters, in that it represents an adaptive or maladaptive practice. To examine the above research question data from 8,475 individuals completing the computerized version of the Postgraduate General Aptitude Test (PAGAT) were analyzed. To determine the extent to which response latency affects person's ability, we used a Multiple Indicators Multiple Causes (MIMIC) model, in which every item in a scale was linked to its corresponding covariate (i.e., item response latency). We ran the MIMIC model within the Item Response Theory (IRT) framework (2-PL model). The results supported the hypothesis that item response latency could provide valuable information for getting more accurate estimations for persons' ability levels. Results indicated that for individuals who invest more time on easy items, their likelihood of success does not improve, most likely because slow and fast responders have significantly different levels of ability (fast responders are of higher ability compared to slow responders). Consequently, investing more time for low ability individuals does not prove to be adaptive. The opposite was found for difficult items: individuals spending more time on difficult items increase their likelihood of success, more likely because they are high achievers (in difficult items individuals who spent more time were of significantly higher ability compared to fast responders). Thus, it appears that there is an interaction between the difficulty of the item and person abilities that explain the effects of response time on likelihood of success. We concluded that accommodating item response latency in a computerized assessment model, can inform test quality and test takers' behavior, and in that way, enhance score measurement accuracy.

What empowers Egyptian women: resources versus social constrains?

Purpose This paper aims to tackle an important question related to women’s economic empowerment in highly patriarchal societies like Egypt. The paper discusses individual, household, wealth and location factors determining women empowerment, as measured by two dimensions: decision-making power and mobility. Design/methodology/approach Using the “Egypt Labor Market Panel Survey” (ELMPS) 2012, a Multiple Indicators Multiple Causes (MIMIC) model was estimated to study the main economic resources and social constraints that determine women empowerment as measured by the power of women over household decisions and her freedom of movement in Egypt. Findings Three key messages could be delivered. First, women’s own economic resources as captured by her employment status are an important source of her empowerment. Second, contrary to theoretical prediction education is not playing its expected role in developing awareness and transforming ideas concerning gender roles in Egypt. Third, the importance of social local context is fundamental for Egyptian women empowerment. Originality/value This study is an attempt to address some of the gaps in the literature for the Egyptian case, where there is a lack in rigorous studies measuring women empowerment and examining its determinates. This is done by first, tackling multiple dimension of women’s empowerment, decision-making inside households and freedom of mobility. Second, using MIMIC model, which is a modeling approach that allows for studying the relations between several causes of a given latent variable, such as “Empowerment” in our case, and a number of its possible indicators, without a directly observable measure of the latent variable. Third, using the most recent set of data; the ELMPS 2012 which has a special focus on women’s resources and agency that permits greater content validity of the multidimensional setup. Forth, the macro level differences in women’s status are tackled through using location dummy variables. Finally, given the important correlation between wealth level and women empowerment, the paper is considered a first attempt to analyze such impact by including a variable that captures the wealth level of the woman’s household as one determinant of empowerment.

The factor structure of the English CBCL DSM-oriented scales in Filipino schoolchildren as reported by bilingual caregivers

ABSTRACTThe Child Behavior Checklist (CBCL) is an assessment tool that has been translated into various languages for use with school-aged children in different cultures. The English form is used in the Philippines, but there is no published evidence on the CBCL’s structural validity when used with Filipino-English bilingual respondents. The study examined the factor structure of the CBCL DSM-oriented scales in a sample of schoolchildren using responses of their bilingual caregivers. Three competing factor models were investigated and confirmatory factor analyses showed that the six-factor model had the best fit. A multiple-indicators multiple-causes model also revealed some predictive relationships with age and gender. The results provided support for the structural validity of the English CBCL DSM-oriented scales when used by Filipino-English bilingual caregivers of Filipino school children.

Impact of error structure misspecification when testing measurement invariance and latent-factor mean difference using MIMIC and multiple-group confirmatory factor analysis.

When multiple groups are compared, the error variance-covariance structure is not always invariant between groups. In this study we investigated the impacts of misspecified error structures on testing measurement invariance and the latent-factor mean difference between groups. A Monte Carlo study was conducted to examine how measurement invariance and latent mean difference tests were affected when heterogeneous error structures were misspecified as being invariant across groups. Multiple-group confirmatory factor analysis (MGCFA) and the multiple-indicator multiple-causes model (MIMIC) were employed in the present study. The rejection rates of both metric and strict invariance in measurement invariance testing, as well as the estimation accuracy and statistical inference of the factor mean difference, were investigated under error structure misspecification. In addition, sensitivity of the model fit indices to error structure misspecification was examined. Overall, misspecification of the error structure affected testing for metric but not scalar invariance. Metric invariance was often rejected, especially when error covariance in one group was ignored. In contrast, MGCFA and MIMIC performed comparatively well at detecting latent-factor mean differences between groups, with acceptable power and well-controlled Type I errors. The practical implications of these findings are discussed, as well as recommendations.