ISSN 0439-755X
CN 11-1911/B


    25 September 2022, Volume 54 Issue 9 Previous Issue    Next Issue

    Reports of Empirical Studies
    For Selected: Toggle Thumbnails
    Reports of Empirical Studies
    The enhancing effect of tracking gesture on visuo-spatial learning
    JIA Xiaoqian, SONG Xiaolei
    2022, 54 (9):  1009-1020.  doi: 10.3724/SP.J.1041.2022.01009
    Abstract ( 1877 )   HTML ( 139 )  
    PDF (1131KB) ( 2696 )   Peer Review Comments

    Visuo-spatial ability is the main component of spatial cognitive ability, which forms the spatial model of external world in human brain and think in mind through the transformation of model paradigm. However, when there are few landmarks or reference objects in an unfamiliar environment, individuals with low visuo-spatial ability are often prone to get lost. So, for these individuals with low visuo-spatial ability, how to solve this problem is worth exploring. Previous researches have shown that gestures help individuals to learn. However, most studies focused on improving the ability of route learning by observing tracking gestures, while there were few studies exploring the effect of generating tracking gestures, and there was no research exploring the role of generating tracking gestures in route learning on 3D maps.
    The route learning task was adopted as the main paradigm in the present study to explore the enhancing effect of generating tracking gestures in the route learning process of 3D map. A total of 53 participants were recruited in Experiment 1 to explore the improving effect of generating tracking gestures on visuo-spatial learning of 2D map and 3D map. In Experiment 2, another 49 participants were recruited to investigate the enhancing mechanism of generating tracking gestures in route learning using occlusion paradigm. Experiment 3 recruited another 53 participants to explore the methods of enhancement of visuo-spatial learning based on tracking gesture.
    The results of Experiment 1 showed that generating tracking gestures significantly improve the performance of visuo-spatial learning in both 2D map and 3D map, and the enhancing effect of generating tracking gesture on route learning of 3D map was higher than that of 2D map. Experiment 2 revealed that visuo-spatial learning was enhanced by the visual information and sensorimotor information provided by tracking gestures. The results of Experiment 3 showed that the enhancing effect of generating self-referential tracking gestures was significantly higher than that of observing others' tracking gestures, and it also proved that enhancing physical presence can improve the effect of visuo-spatial learning.
    In summary, these results suggest that (1)generating tracking gestures can enhance visuo-spatial learning; (2)visual information and sensorimotor information provided by tracking gestures play important roles in visuo-spatial learning, which confirms image maintenance theory and further complements the embodied embodied-external cognitive perspective of gesture; (3)increasing physical presence can enhance visuo-spatial learning, which validates the embodied-external cognitive perspective of gesture; (4)compared with observing other's tracking gestures, generating self-referential tracking gestures has a better performance, that is, generating self-referential tracking gestures is a more effective way to improve visuo-spatial learning.

    Figures and Tables | References | Related Articles | Metrics
    The cross-domain influence of tonal categorical perception and tonal complexity on musical pitch perception
    LI Xianzhuo, XIAO Rong, LIANG Dandan
    2022, 54 (9):  1021-1030.  doi: 10.3724/SP.J.1041.2022.01021
    Abstract ( 629 )   HTML ( 46 )  
    PDF (1479KB) ( 815 )   Peer Review Comments

    Pitch is a fundamental acoustic property shared by both language and music. However, there are different views on the processing of tonal pitch and musical pitch. Some studies support the “modularity view”, suggesting that tonal pitch and musical pitch are represented separately. In contrast, more studies support the “shared domain-general view”, implying that there are commonalities in the processing of tonal pitch and musical pitch based on their physical similarities. The existing studies have mostly focused on domain-general pitch transfer; nonetheless, they have not considered domain specificity of the tonal pitch as a linguistic element. Hence, it remains unclear whether domain specificity factors, such as categorical perception and complexity of different tonal language experiences on speakers’ pitch processing, play a role in musical pitch perception.
    To address the above issues, ninety participants were involved in the experiment, including native speakers of Mandarin Chinese (with a relatively simple pitch category), native speakers of Vietnamese (with a relatively complex pitch category), and native speakers of Russian (nonpitched control group). A 3 (group: Vietnamese vs. Chinese vs. Russian) × 2 (stimulus type: speech vs. music) between-and-within-subjects design was used. A continuum from [i?] (closed for Yinping in Chinese and Transverse in Vietnamese) to [i??] (closed for Yangping in Chinese and Acute in Vietnamese) and its musical counterpart was constructed. Participants were first tested in an ABX identification task to determine whether the stimulus X was similar to A or B and then an AX discrimination task to decide whether the two stimuli in the pair were identical. The experiment was performed in E-prime 3.0.
    The results showed that (1) both the Chinese and Vietnamese groups showed categorical perception for language and music pitches. There were no differences between language and music stimuli in category boundary width, within-category discrimination rate, between-category discrimination rate, or discrimination peak for either group. The Russian group’s identification curve did not show abrupt shifts, and their discrimination curve was relatively flat with multiple peaks, which indicates a continuous pattern significantly different from the two tonal language groups. (2) There were no significant differences between the Vietnamese group and the Chinese group in either the within-category discrimination rate or the between-category discrimination rate.
    The experimental results suggest that at the behavioral level, the pattern of native tonal categorical perception can transfer to musical pitch perception, but tonal complexity does not facilitate cross-domain musical pitch perception. The findings of this study support the “shared domain-general view” in terms of the influence of language on musical pitch processing.

    Figures and Tables | References | Related Articles | Metrics
    Efficacy of suicide ideation classification based on pain avoidance and the EEG characteristics under self-referential punishment
    SUN Fang, SONG Wei, WEN Xiaotong, LI Huanhuan, OUYANG Lisheng, WEI Shijie
    2022, 54 (9):  1031-1047.  doi: 10.3724/SP.J.1041.2022.01031
    Abstract ( 1555 )   HTML ( 123 )  
    PDF (8661KB) ( 2614 )   Peer Review Comments

    Depressed students are at high-risk for suicide. Psychological pain, especially pain avoidance, was a more robust predictor for suicide ideation than depression at the behavioral level. Due to suicide as a complex classification model, machine learning algorisms applied to integrate behavioral data and neural characteristic can advance suicide prediction, and the accuracy of multimodality features is superior than clinical interview. The present study aimed to integrate data-driven machine learning algorisms and the three-dimensional psychological pain model to figure out the optimal features in the prediction of suicide ideation.
    Seventy-seven college students were recruited by advertisement and divided into three groups: depressed group with high levels of suicide ideation (HSI, n = 25), depressed group with low levels of suicide ideation (LSI, n = 20), and healthy controls (HC, n = 32). All participants completed the three-dimensional psychological pain scale (TDPPS), Beck depression inventory-I (BDI), Beck suicide ideation inventory (BSI), and the self-referential affective incentive delay task (SAID). The value of support vector based on machine-recursive feature elimination (RFE-SVM) algorithm applied to combine the scale scores, resting state and punitive-related EEG components for feature ranking in a nonlinear way.
    Results showed that: (1) Scores of pain avoidance in the HSI was higher than the LSI group. (2) The multimodal psychological pain-based model for suicide ideation classification (Accuracy = 85.66%, Precision = 0.82, Recall = 0.73, AUC = 0.92) was sufficient and superior than the EEG single-modal model. Importantly, the pain avoidance and BDI scores ranked the top two features in the classification model of suicide ideation, whereas painful feeling and pain arousal subscale scores ranked the top two features in the classification model of depression. The EEG optimal features of overlap in the pain avoidance and suicide ideation classification models were the LPP and target-P3 under self-referential punitive conditions. (3) The powers of delta and beta band were negatively correlated with the BSI-W and pain avoidance subscale scores. The FRN amplitude under other- and self-referential punitive conditions were negatively corelated with the pain avoidance subscale scores. In the HSI group, power of delta elicited by positive feedback under self-referential conditions was significantly lower than those under other-referential conditions. In the HSI group, the amplitude of LPP in other-referential punitive conditions was higher than those under reward and neutral conditions, whereas in the LSI group, the amplitude of LPP under self-referential punitive conditions was higher than that under neutral conditions.
    As a pilot study, the current study provided a support for the prominent role of pain avoidance and its related neuroelectrophysiological correlates in the prediction of suicide. The clinical significance of these results will be discussed.

    Figures and Tables | References | Related Articles | Metrics
    The covariant relationship between adolescent friendship networks and bullying: A longitudinal social network analysis
    ZHANG Yunyun, ZHANG Qiwen, ZHANG Libin, REN Ping, QIN Xingna, CHANG Ruisheng
    2022, 54 (9):  1048-1058.  doi: 10.3724/SP.J.1041.2022.01048
    Abstract ( 2010 )   HTML ( 153 )  
    PDF (577KB) ( 2957 )   Peer Review Comments

    Bullying is a group process that involves multiple behaviors and multiagent interactions. However, few studies have focused on the development of bullying-related behaviors in the context of peer group interaction and relationship change, particularly in friendship networks. Although social support theory suggests that positive friendships act as a buffer, previous research has found that victims have difficulty making friends. Furthermore, recent research on friendship networks has indicated that the potential buffering role of friends may depend on the extent to which friends engage in certain behaviors. It is thus crucial to determine how victims shape their friendship networks and the protection mechanisms or detrimental effects associated with victims that such networks can trigger. Accordingly, the aim of this study was to longitudinally examine the selection and influence effects of victimization in dynamic classroom friendship networks using a sample of Chinese adolescents. We expected that (1) while peers tend to avoid selecting victims as friends (H1a: peer avoidance effect), victims tend to select other victimized youth as friends (H1b: peer selection effects), and that (2) the more friends that an adolescent has, the less he or she is victimized (H2a: peer protective effect), whereas one's victimization increases as a result of one’s friends being victimized (H2b: peer influence effect).
    A total of 1406 eighth-grade students in public middle schools from Central China were recruited to participate in three assessments, with an interval of six months. Peer nomination was applied to evaluate friendship and victimization, with a maximum of five nominations allowed from classmates who fit the description. Capitalizing on longitudinal social network analysis (SIENA) to disentangle selection and influence processes, this study focuses on the role of friendship in adolescents’ peer victimization after controlling for structural characteristics of the network (e.g., transitivity and reciprocity) in friendship network dynamics.
    It was found that (1) while peers avoided selecting victims as friends (peer avoidance effect), victims tended to select other victims as friends (peer selection effect); (2) whether peers nominated others as friends was not influenced by their level of victimization; (3) having more friends decreased one’s level of victimization (peer protection effect), while being friends with a victim increased one’s possibility of being victimized over time (peer influence effect); (4) the higher the victimization level of an adolescent was, the higher his or her victimization level gradually rose over time (enhanced effect of victimization); and (5) the level of victimization decreased over time among girls compared to that of boys. These results revealed the peer selection and influence effects on the covariation of friendship networks and victimization among Chinese adolescents and demonstrated the peer “contagion” that is associated with victimization, which provides insights for interventions for school bullying.

    Figures and Tables | References | Related Articles | Metrics
    The effect of group identity shifting on impression updating in older adults: The mediating role of common ingroup identity
    WEN Fangfang, KE Wenlin, HE Saifei, ZUO Bin, LI Lanxin, MA Shuhan, WANG Jing
    2022, 54 (9):  1059-1075.  doi: 10.3724/SP.J.1041.2022.01059
    Abstract ( 1488 )   HTML ( 107 )  
    PDF (881KB) ( 1783 )   Peer Review Comments

    In the field of impression updating, most researchers focus on the impression updating of individual targets, while few researches focus on the impression updating of groups. However, the intervention of prejudice and conflict between groups has always been a hot issue to be solved in the field of social psychology. Since group identity is the basis for impression evaluation of groups, based on the perspective of Common Ingroup Identity Model and social categorization, changing group identity to improve common ingroup identity provides a feasible “change makes sense” intervention path for the impression update of the target group. Considering that the large elderly population has become an important part of the world population, it is of great practical significance to evaluate the elderly population positively. Based on this, the present study manipulated group identity shifting at both explicit and implicit levels through “minimal group recategorization paradigm”. The purpose of this study is to explore the influence of identity shifting on the impression updating of the elderly group and the role of common ingroup identity in it, so as to explore a method with low restriction, simple operation and obvious effect to improve out-group impression evaluation.
    In preliminary experiment, 119 college students participated in the psychology class as subjects, to examine the effect of group identity shifting manipulated by the minimal group recategorization paradigm on the updating of minimal in-group/out-group impressions. In Experiment 1A, 98 college students were recruited as subjects. The aim of Experiment 1A was to use the “minimal group recategorization paradigm” to manipulate the shifting of group identity, examine the change of young individuals’ perception of warmth and competence towards the elderly group, and examine the effect of group identity shifting on the impression updating of the elderly at the explicit level. In Experiment 1B, we recruited 35 college students as subjects to explore whether the impression evaluation of the elderly can be effectively changed at the implicit level by using Go/ No-Go association task. In Experiment 2, the effect of identity shifting was tested on both explicit and implicit levels. At the implicit level, a Single Category Implicit Association Test with higher application rate and wider application scope was used, and an integrated perspective of warmth, competence and stereotype trait words was used to measure the impression evaluation. At the same time, the influence of identity shifting on common ingroup identity was also measured to explore the role of common ingroup identity in the impression updating of the elderly.
    The results of pre-experiment found that shifting group identity could effectively update individual’s impression of out-group. Specifically, the evaluation of out-group in the changed group was significantly improved compared with that in the unchanged group. The results of experiment 1A showed that at the explicit level, shifting group identity could affect the impression updating of the elderly, and the second evaluation of the elderly has been significantly improved, especially in the competence dimension. The results of Experiment 1B showed that at the implicit level, manipulation of group identity shifting could not effectively improve the impression evaluation of the elderly. In experiment 2, it was found that at the explicit level, common ingroup identity played a completely mediating role in the effect of identity shifting on impression improvement. At the implicit level, consistent with experiment 1B, identity shifting did not change subjects' implicit impression evaluation of the elderly.
    In conclusion, minimal group recategorization paradigm can shift individual group identity through two classifications and improve impression evaluation of out-group, which is an effective tool to change group identity and alleviate intergroup bias. At the explicit level, common ingroup identity played a mediating role, but at the implicit level, no intervention effect was found. These findings have created a new recategorization method for enhancing common ingroup identity, and opened up a simple and effective method for intergroup prejudice intervention. Moreover, the application of minimal group paradigm is no longer the static application, but more dynamic, ecological significance and realistic value.

    Figures and Tables | References | Related Articles | Metrics
    Algorithmic discrimination causes less desire for moral punishment than human discrimination
    XU Liying, YU Feng, PENG Kaiping
    2022, 54 (9):  1076-1092.  doi: 10.3724/SP.J.1041.2022.01076
    Abstract ( 1153 )   HTML ( 80 )  
    PDF (896KB) ( 1461 )   Peer Review Comments

    The application of algorithms is believed to contribute to reducing discrimination in human decision-making, but algorithmic discrimination still exists in real life. So is there a difference between folk responses to human discrimination and algorithmic discrimination? Previous research has found that people's moral outrage at algorithmic discrimination is less than that at human discrimination. Few studies, however, have investigated people's behavioral tendency towards algorithmic discrimination and human discrimination, especially whether there is a difference in their desire for moral punishment. Therefore, the present study aimed at comparing people's desire to punish algorithmic discrimination and human discrimination as well as finding the underlying mechanism and boundary conditions behind the possible difference.
    To achieve the research objectives, six experiments were conducted, which involved various kinds of discrimination in daily life, including gender discrimination, educational background discrimination, ethnic discrimination and age discrimination. In experiment 1 and 2, participants were randomly assigned to two conditions (discrimination: algorithm vs. human), and their desire for moral punishment was measured. Additionally, the mediating role of free will belief was tested in experiment 2. To demonstrate the robustness of our findings, the underlying mechanism (i.e., free will belief) was further examined in experiment 3 and 4. Experiment 3 was a 2 (discrimination: algorithm vs. human) × 2 (belief in free will: high vs. low) between-subject design, and experiment 4 was a single-factor (discrimination: human vs. algorithm with free will vs. algorithm without free will) between-subject design. Experiment 5 and 6 were conducted to test the moderating role of anthropomorphism. Specifically, participants’ tendency to anthropomorphize was measured in experiment 5, and the anthropomorphism of algorithm was manipulated in experiment 6.
    As predicted, the present research found that compared with human discrimination, people have less desire to punish algorithmic discrimination. And the robustness of this result was demonstrated by the diversity of our stimuli and samples. In addition, we found that free will belief played a mediating role in the effect of discrimination (algorithm vs. human) on the desire to punish. That is to say, the reason why people had less desire to punish when facing algorithm discrimination was that they thought algorithms had less free will than humans. Finally, the results also demonstrated the moderating effect of anthropomorphism.
    These results enrich literature regarding algorithm discrimination as well as moral punishment from the perspective of social psychology. First, this research explored people's behavioral tendency towards algorithmic discrimination by focusing on the desire for moral punishment, which contributes to a better understanding of people's responses to algorithmic discrimination. Second, the results are consistent with previous studies on people’s mind perception of artificial intelligence. Third, it adds evidence that free will has a significant impact on moral punishment.

    Figures and Tables | References | Related Articles | Metrics
    The impact of ethical leadership prototype on the effectiveness of ethical leadership: The mediating role of elevation
    XING Zhijie, HE Wei, ZHANG Zhengtang, JIANG Xuting
    2022, 54 (9):  1093-1105.  doi: 10.3724/SP.J.1041.2022.01093
    Abstract ( 848 )   HTML ( 68 )  
    PDF (632KB) ( 1356 )   Peer Review Comments

    Ethical leadership (EL) is a positive leadership style demonstrated to be effective in promoting employee performance. However, previous research has mainly focused on the cognitive perspective to explain the positive effects of ethical leadership on employee outcomes, and the role of emotion has received little attention. In this research, we advance this stream of research by taking into account employees’ implicit beliefs of ethical leadership (i.e., ethical leadership prototype, ELP). Specifically, drawing from the theoretical model of discrete positive emotions in the workplace and implicit leadership theories, this paper theorized and tested the mediating role of elevation on the relationship between ethical leadership and employee task performance and organizational citizenship behavior (OCB) moderated by employee ELP.
    We tested our hypotheses in a multi-source, multi-wave field study (Study 1). In Study 1, we tested our full theoretical model (i.e., H1 to H4) by collecting multi-wave and multi-source data from 193 leader-employee dyads. Empirical results in Study 1 only demonstrated the positive relationship between EL and employee elevation (H1) and the indirect relationship between EL and employee OCB via employee elevation (H2b).
    To replicate these empirical findings and test the moderating role of employee ELP further, we conducted two vignette experiments (Study 2 and 3). In Study 2 and Study 3, we recruited 400 and 200 participants from two online survey panels (i.e., the Sojump and the Credamo) and manipulated ethical vs. unethical leadership (Study 2) and high vs. low ethical leadership (Study 3). Experimental results in both studies replicated the positive relationship between EL and employee elevation (H1) and the indirect relationship between EL and OCB via elevation (H2b); in addition, the moderating effects of employee ELP on the relationship between EL and elevation (H3) and the indirect relationship between EL and OCB via elevation (H4b) were demonstrated.
    This research makes several theoretical contributions. First, this research contributes to our understanding of the process through which EL influences employee outcomes by identifying employee ELP as a crucial contingency. Second, by demonstrating the mediating mechanism of the positive moral emotion of elevation (with general positive affect being controlled), this research provides a finer-grained understanding of the affect-based mechanism for the effectiveness of EL. Third, by investigating when EL activates employee elevation, this paper enriches the knowledge of the antecedents of felt elevation. Finally, this paper extends implicit leadership theory by applying it to a specific leadership domain and highlighting the crucial role of ELP.

    Figures and Tables | References | Related Articles | Metrics
    Can you perceive my efforts? The impact of social status on consumers’ preferences for complexity
    CHEN Zengxiang, HE Yun, LI Xiao, WANG Lin
    2022, 54 (9):  1106-1121.  doi: 10.3724/SP.J.1041.2022.01106
    Abstract ( 2354 )   HTML ( 164 )  
    PDF (822KB) ( 3223 )   Peer Review Comments

    Aesthetic is the first medium of a product in the era of “appearance economy.” Previous studies have explored the effects of aesthetic elements on consumer behavior, including color, packaging transparency, size, logo shape, and so on. Unlike existing studies, this research discusses the overall dimension of design, that is, the degree of complexity of the design. Specifically, we propose that people’s subjective social status affects their preference for complexity in design. Meanwhile, people with low social status are more likely to prefer a complex packaging design than those with high social status.
    We conducted five experiments to test the hypotheses. Experiment 1 used 2 (social status: low vs. high) × 2 (product design: simple vs. complex) between-subjects design with 198 adults (female = 128) participating in the experiment. The results showed that people with relatively low social status have higher evaluations of complex packaging designs (vs. simple designs). At the same time, the level of complexity did not influence how people with high social status evaluate products. Using a more rigorous method to manipulate the social status, Experiment 2 employed a one-factor two-level (social status: low vs. high) between-subjects design with 134 adults (female = 97). After manipulating social statuses, participants reported their preferences between complex and simple packaging designs. The result confirmed that people with low social status prefer products with complex designs (vs. simple designs).
    Similar to Experiment 2, Experiment 3 (211 participants, female = 151) also used a one-way design (social status: low vs. high vs. control) but added a control group as the baseline group for comparison. The results confirmed that a low social status could drive preference for complexity. Furthermore, the results showed that people with low social status prefer complex packing design more than those in control and high social status groups. At the same time, no significant difference was observed in the preference between the high social status and control groups. Experiment 3 also found that consumers’ perceived effort mediated the above effect. Specifically, low social status individuals (vs. individuals in high social status vs. individuals in the control group) perceived that complex packing products entailed more effort from producers. Hence, they preferred such products.
    Experiment 4 was a pre-registration experiment (200 participants, female = 151). The experimental design and procedure were the same as in Experiment 2. The results also showed that people with low social status preferred complex packaging designs and believed these products had a higher value. Compared with high social status individuals, low social status individuals perceived that producers spent more effort on a product if it had a complex packaging design. Thus, these consumers perceived higher product value, as well as higher preference toward the product (sequential mediation: model 6).
    Experiment 5 tested the mediation effect by moderators. We proposed that the impact of social status only existed in consumers who believed in the value of effort. Experiment 5 used a 2 (social status: low vs. high) × 2 (product design: simple vs. complex) × 2 (belief in effort value: low vs. high) between-subjects design (346 participants, female = 208). The ANOVA revealed a three-way interaction. Furthermore, the Johnson-Neyman analysis found that the preference for complex design among low social status individuals only existed among those who believed in the value of effort. The results of this interaction again suggested that the preference of low social status individuals for complexity stems from their perception of the producer’s effort in making these products.
    This study has the following theoretical contributions. First, it brings the perspective of subjective social status to the study of consumer aesthetics. Second, it provides a new mechanism for the role of social status, i.e., social status affects people’s behavior by influencing their perceived importance of effort. Previous explanations of the behavior of low social status individuals have tended to be from a compensatory psychological perspective. In contrast, the current research takes the perspective of effort to demonstrate that low social status individuals value effort, which influences their behavior. Third, this research also contributes to the study of consumer effort, as this study finds that simply changing the level of complexity of packaging design can change people’s perceptions of producer effort.

    Figures and Tables | References | Related Articles | Metrics
    Detection of aberrant response patterns using a residual-based statistic in testing with polytomous items
    TONG Hao, YU Xiaofeng, QIN Chunying, PENG Yafeng, ZHONG Xiaoyuan
    2022, 54 (9):  1122-1136.  doi: 10.3724/SP.J.1041.2022.01122
    Abstract ( 271 )   HTML ( 6 )  
    PDF (3463KB) ( 928 )   Peer Review Comments

    Tests are widely used in educational measurement and psychometrics, and the examinee’s aberrant responses will affect the estimation of their abilities. These examinees with aberrant responses should not be treated with conventional methods, the important thing is to accurately screen them out of the normal group. To achieve this, a common method is to construct person-fit statistics to detect whether the response patterns fit their estimated abilities.
    In this study, a residual-based person-fit statistic R was proposed, which can be applied to both dichotomous or polytomous IRT models. The construction of R is based on a weighted residual between the observed response and the expected response. By accumulating the weighted residuals, the goodness of fit can be calculated and compared with a specific critical value to determine whether an examinee is aberrant or not. Given that tests with polytomous items can provide more information, polytomously scored items are being increasingly popular in educational measurement and psychometrics. The ability of R statistic to detect aberrant response patterns under the graded response model was mainly considered in this article.
    An existing polytomous person-ft statistic lzp was also introduced in its outstanding standardized form and superior power. In the first study, a simulation study was conducted to generate the empirical distribution of R statistic and lzp. R statistic is an accumulation of weighted residuals, showing a positive skew distribution; lzp shows a negative skew distribution when the test is less than 80 items. Both of them differ from the standard normal distribution, It is necessary to set critical value according to the type 1 error, using it to distinguish whether each respondent's response pattern is fitted. In the second study, examinees with different aberrant behaviors (e.g., Cheaters, Lucky guessers, Random respondents, Careless respondents, Creative respondents and Mixed) under different test length conditions were simulated, and the detection rate as well as area under curve (AUC) were used to compare the effectiveness of the two person-fit statistics. The results show that the R statistic has a better detection rate than lzp when the aberrant behavior affects only a few items or the aberrant behavior is cheating or guessing. When the aberrant behavior covers plenty of items, lzp is slightly better than R statistic. Then, an empirical study was also conducted to show the power of R statistic.
    Both of the R statistic and the lzp have their own pros and cons, so we may combine them in the future person-fit studies. The R statistic has a better detection rate under certain conditions compared to the lzp, especially when cheating and lucky guessing happened. Considering that cheating and guessing behaviors of low-ability examinees are more preferred in many aberrant test behaviors, the R statistic is worthy of further research and exploration in real-world applications.

    Figures and Tables | References | Related Articles | Metrics
    Nonparametric cognitive diagnostic computerized adaptive testing using multiple-choice option information
    SUN Xiaojian, GUO Lei
    2022, 54 (9):  1137-1150.  doi: 10.3724/SP.J.1041.2022.01137
    Abstract ( 328 )   HTML ( 23 )  
    PDF (887KB) ( 351 )   Peer Review Comments

    Most existing cognitive diagnostic computerized adaptive testing (CD-CAT) item selection methods ignore the diagnostic information that distractors provide for multiple-choice (MC) items. Consequently, some useful information is missed and resources are wasted. To overcome this, researchers proposed the Jensen-Shannon divergence (JSD) strategy to select items with the MC-DINA model. However, the JSD strategy needs large samples to obtain reliable estimates of the item parameters before the formal test, and this could compromise the items in the bank. By contrast, the nonparametric method does not require any parameter calibration before the formal test and can be used in small educational programs.
    The current study proposes two nonparametric item selection methods (i.e., HDDmc and JDDmc) for CD-CAT with MC items as well as two termination rules (i.e., MR and DR) for variable-length CD-CAT with MC items. Two simulation studies were conducted to examine the performance of these nonparametric item selection methods and termination rules.
    The first study examined the performance of the HDDmc and JDDmc with fixed-length CD-CAT. In this study, six factors were manipulated: the number of attributes (K = 4 vs. 6), the structure of the Q-matrix (simple vs. complex), the quality of the item bank (high vs. low vs. mixed), the distribution of the attribute profile (multivariate normal threshold model vs. discrete uniform distribution), the test length (two vs. three vs. four times of K), and the item selection methods (HDDmc vs. JDDmc vs. JSD). Of these, item selection method was the within-group variable, and the rest were between-group variables. The results showed that: (1) the HDDmc and JDDmc produced higher attribute pattern matched ratios (PMRs) than the JSD method for most conditions; (2) the HDDmc and JDDmc produced similar PMRs for all conditions; (3) the HDDmc and JDDmc produced more even distributions of item exposure than the JSD method.
    The second simulation study investigated the performance of the MR and DR with variable-length CD-CAT. Six factors were also manipulated in this study: the settings for the number of attributes, the structure of the Q-matrix, the quality of the item bank, and the distribution of the attribute profile were the same as in the first study; the other two factors were termination rules (MR, DR, D1, and D3) and item selection methods (HDDmc and JDDmc). Again, the first four were between-group variables, while termination rules and item selection methods were within-group variables. The results showed that: (1) the HDDmc and JDDmc yielded higher PMRs for MR and DR rules than for the D1 and D3 rules; (2) the HDDmc and JDDmc yielded longer test lengths for MR and DR rules than for the D1 and D3 rules, especially for the JDD rule.
    In sum, both nonparametric item selection methods and the two new termination rules proved appropriate for CD-CAT with MC items, which means they can be used to balance the trade-off between measurement accuracy and item exposure rate.

    Figures and Tables | References | Related Articles | Metrics