Fast times and easy questions: the effects of age,experience and question complexity on web survey response times

This paper examines response times (RT) to survey questions. Cognitive psychologists have long relied on response times to study cognitive processes but response time data have only recently received attention from survey researchers. To date, most of the studies on response times in surveys have treated response times either as a predictor or as a proxy measure for some other variable (e.g. attitude accessibility) of greater interest. As a result, response times have not been the main focus of the research. Focusing on the nature and determinants of response times, this paper examines variables that affect how long it takes respondents to answer questions in web surveys. Using the survey response model proposed by Tourangeau, Rips, and Rasinski (2000), we include both item‐level characteristics and respondent‐level characteristics thought to affect response times in a two‐level cross‐classified model. Much of the time spent on processing the questions involves reading and interpreting them. The results from the cross‐classified models indicate that response times are affected by question characteristics such as the total number of clauses and the number of words per clause that probably reflect reading times. In addition, response times are also affected by the number and type of answer categories, and the location of the question within the questionnaire, as well as respondent characteristics such as age, education and experience with the Internet and with completing web surveys. Aside from their fixed effects on response times, respondent‐level characteristics (such as age) are shown to vary randomly over questions and effects of question‐level characteristics (such as types of questions and response scales) vary randomly over respondents. Copyright © 2007 John Wiley & Sons, Ltd.     

Cognitive diagnosis modelling incorporating item response times

To provide more refined diagnostic feedback with collateral information in item response times (RTs), this study proposed joint modelling of attributes and response speed using item responses and RTs simultaneously for cognitive diagnosis. For illustration, an extended deterministic input, noisy ‘and’ gate (DINA) model was proposed for joint modelling of responses and RTs. Model parameter estimation was explored using the Bayesian Markov chain Monte Carlo (MCMC) method. The PISA 2012 computer-based mathematics data were analysed first. These real data estimates were treated as true values in a subsequent simulation study. A follow-up simulation study with ideal testing conditions was conducted as well to further evaluate model parameter recovery. The results indicated that model parameters could be well recovered using the MCMC approach. Further, incorporating RTs into the DINA model would improve attribute and profile correct classification rates and result in more accurate and precise estimation of the model parameters.     

Word naming times and psycholinguistic norms for Italian nouns

The present study describes normative measures for 626 Italian simple nouns. The database (LEXVAR. XLS) is freely available for down-loading on the Web site http://wwwistc.ip.rm.cnr.it/material/ database/. For each of the 626 nouns, values for the following variables are reported: age of acquisition, familiarity, imageability, concreteness, adult written frequency, child written frequency, adult spoken frequency, number of orthographic neighbors, mean bigram frequency, length in syllables, and length in letters. A classification of lexical stress and of the type of word-initial phoneme is also provided. The intercorrelations among the variables, a factor analysis, and the effects of variables and of the extracted factors on word naming are reported. Naming latencies were affected primarily by a factor including word length and neighborhood size and by a word frequency factor. Neither a semantic factor including imageability, concreteness, and age of acquisition nor a factor defined by mean bigram frequency had significant effects on pronunciation times. These results hold for a language with shallow orthography, like Italian, for which lexical nonsemantic properties have been shown to affect reading aloud. These norms are useful in a variety of research areas involving the manipulation and control of stimulus attributes.     

Performance structures: A psycholinguistic and linguistic appraisal

Two lines of research—one in psycholinguistics and one in linguistics—are combined to deal with a long-standing problem in both fields: why the “performance structures” of sentences (structures based on experimental data, such as pausing and parsing values) are not fully accountable for by linguistic theories of phrase structure. Two psycholinguistic algorithms that have been used to predict these structures are described and their limitations are examined. A third algorithm, based on the prosodic structures of sentences is then proposed and shown to be a far better predictor of performance structures. It is argued that the experimental data reflect aspects of the linguistic cognitive capacity, and that, in turn, linguistic theory can offer an illuminating account of the data. The prosodic model is shown to have a wider domain of application than temporal organization per se, accounting for parsing judgments as well as pausing performance, and reflecting aspects of syntactic and semantic structure as well as purely prosodic structure. Finally, the algorithm is discussed in light of language processing.     

Cognitive mechanisms underlying lying to questions: response time as a cue to deception

In this article, a model of the cognitive processes involved in generating deceptive answers to questions is proposed. Partly based on recent accounts of discourse processing, it posits two central processes. One is the decision to lie followed by the construction of a lie. Each should add to response time compared to answering honestly. In Experiment 1, the construction component was assessed; 92 adults were randomly assigned to one of two conditions where they answered personal questions either deceptively or truthfully. Constructing a lie added reliably to response time. Think‐aloud and correlational data support features of the model. In Experiment 2, 121 adults were randomly assigned to one of three conditions: either to tell the truth, lie, or lie when asked a question that probed sensitive information. Replicating and extending Experiment 1, the decision to lie added reliably to response time in the case of open‐ended questions (questions that elicit more than two possible answers). Based on between‐ and within‐subjects comparisons, results suggest that response time is a cue to deception. Recommendations are made for future research. Copyright © 2003 John Wiley & Sons, Ltd.     

Patterns of paralexia: A psycholinguistic approach

A sample of the literature on acquired dyslexia is reviewed with special reference to the nature of paralexic errors. Linguistic studies of six cases of dyslexia are presented, with appropriate neurological and psychological details; there are two cases of each of three hypothesized types of impairment, visual dyslyxia, surface (grapheme-phoneme) dyslexia, and deep (syntactico-semantic) dyslexia. Finally, a scheme for acquired dyslexia is proposed and related to the error patterns observed in normal adult readers and in children learning to read; our approach to lexical look-up is one that the diagram makers may have found sympathique.This is a revised version of paper presented at the International Neuropsychology Symposium, Engelberg, Switzerland, June 1971.     

Concreteness norms for 1,659 French words: Relationships with other psycholinguistic variables and word recognition times

Words that correspond to a potential sensory experience—concrete words—have long been found to possess a processing advantage over abstract words in various lexical tasks. We collected norms of concreteness for a set of 1,659 French words, together with other psycholinguistic norms that were not available for these words—context availability, emotional valence, and arousal—but which are important if we are to achieve a better understanding of the meaning of concreteness effects. We then investigated the relationships of concreteness with these newly collected variables, together with other psycholinguistic variables that were already available for this set of words (e.g., imageability, age of acquisition, and sensory experience ratings). Finally, thanks to the variety of psychological norms available for this set of words, we decided to test further the embodied account of concreteness effects in visual-word recognition, championed by Kousta, Vigliocco, Vinson, Andrews, and Del Campo (Journal of Experimental Psychology: General, 140, 14–34, 2011). Similarly, we investigated the influences of concreteness in three word recognition tasks—lexical decision, progressive demasking, and word naming—using a multiple regression approach, based on the reaction times available in Chronolex (Ferrand, Brysbaert, Keuleers, New, Bonin, Méot, Pallier, Frontiers in Psychology, 2; 306, 2011). The norms can be downloaded as supplementary material provided with this article.     

A mixture hierarchical model for response times and response accuracy

In real testing, examinees may manifest different types of test‐taking behaviours. In this paper we focus on two types that appear to be among the more frequently occurring behaviours – solution behaviour and rapid guessing behaviour. Rapid guessing usually happens in high‐stakes tests when there is insufficient time, and in low‐stakes tests when there is lack of effort. These two qualitatively different test‐taking behaviours, if ignored, will lead to violation of the local independence assumption and, as a result, yield biased item/person parameter estimation. We propose a mixture hierarchical model to account for differences among item responses and response time patterns arising from these two behaviours. The model is also able to identify the specific behaviour an examinee engages in when answering an item. A Monte Carlo expectation maximization algorithm is proposed for model calibration. A simulation study shows that the new model yields more accurate item and person parameter estimates than a non‐mixture model when the data indeed come from two types of behaviour. The model also fits real, high‐stakes test data better than a non‐mixture model, and therefore the new model can better identify the underlying test‐taking behaviour an examinee engages in on a certain item.     

Cognitive and psycholinguistic skills of adults who are functionally illiterate: Current state of research and implications for adult education

The purpose of this article is to review empirical studies that have investigated the cognitive and language skills of adults who are functionally illiterate (A‐IL), that is, individuals whose literacy skills are too low to meet the literacy demands of society, even though they have attended school and are not illiterate in the strictest sense of the term. Three questions are addressed: (1) What are the types of reading and spelling difficulties that characterize adults who are A‐IL? (2) Are their written language difficulties related to cognitive or language impairments which could explain their failure to learn? (3) What are the implications of these psychological findings for adult education? Copyright © 2010 John Wiley & Sons, Ltd.     

Word naming and psycholinguistic norms: Chinese

In this article, we present normative data for 2,423 Chinese single-character words. For each word, we report values for the following 15 variables: word frequency, cumulative frequency, homophone density, phonological frequency, age of learning, age of acquisition, number of word formations, number of meanings, number of components, number of strokes, familiarity, concreteness, imageability, regularity, and initial phoneme. To validate the norms, we collected word-naming latencies. Factor analysis and multiple regression analysis show that naming latencies of Chinese single-character words are predicted by frequency, semantics, visual features, and consistency, but not by phonology. These analyses show distinct patterns in word naming between Chinese and alphabetic languages and demonstrate the utility of normative data in the study of nonalphabetic orthographic processing.     

N-Watch: A program for deriving neighborhood size and other psycholinguistic statistics

This article describes a Windows program that enables users to obtain a broad range of statistics concerning the properties of word and nonword stimuli, including measures of word frequency, orthographic similarity, orthographic and phonological structure, age of acquisition, and imageability. It is designed for use by researchers in psycholinguistics, particularly those concerned with recognition of isolated words. The program computes measures of orthographic similarity on line, either with respect to a default vocabulary of 30,605 words or to a vocabulary specified by the user. In addition to providing standard orthographic neighborhood measures, the program can be used to obtain information about other forms of orthographic similarity, such as transposed-letter similarity and embedded-word similarity. It is available, free of charge, from the following Web site: http://www.maccs.mq.edu.au/≈colin/N-Watch/.     

The semantic structure of vision verbs: A psycholinguistic investigation of Italian

The aim of the present study was to investigate the semantic dimensions that contribute to the mental representation of vision verbs in Italian. In Experiment 1, a group of experts (psychology faculty members) was asked to list the semantic attributes necessary for describing a set of Italian vision verbs (Phase 1) and a group of students rated the extent to which each of these attributes was appropriate to characterise these verbs (Phase 2). In Experiment 2, the conceptual organisation of vision verbs was assessed through Multidimensional Scaling (MDS) using direct judgements of pairwise similarity between vision verbs (e.g., “vedere-guardare”). The results show that six of the attributes generated in Phase 1 are useful for describing the pattern of similarity ratings obtained in Experiment 2, namely attention, certainty, cognitive content, control, duration, and intentionality.     

Family burden of compulsive hoarding: results of an internet survey

Compulsive hoarding, the acquisition of and failure to discard large numbers of possessions, is associated with substantial health risk, impairment in functioning, and economic burden. Despite clear indications that hoarding has a detrimental effect on people living with or near someone with a hoarding problem, no empirical research has examined these harmful effects. The aim of the present study was to examine the burden of hoarding on family members. Six hundred sixty-five family informants who reported having a family member or friend with hoarding behaviors completed an internet-based survey. Living with an individual who hoards during childhood was associated with elevated reports of childhood distress and family strain. Family members reported high levels of patient rejection attitudes, suggesting high levels of family frustration and hostility. Rejecting attitudes were predicted by severity of hoarding symptoms, the individual's perceived lack of insight into the behavior, and having lived in a cluttered environment during childhood. These results suggest that compulsive hoarding adversely impacts not only the hoarding individual, but also those living with them.     

A study of syntactic processing in aphasia I: behavioral (psycholinguistic) aspects

This paper presents the results of a study of syntactically based comprehension in aphasic patients. We studied 42 patients with aphasia secondary to left hemisphere strokes and 25 control participants. We measured off-line, end-of-sentence, performance (accuracy and reaction time) in two tasks that require comprehension--enactment and sentence-picture matching--and in grammaticality judgment, with whole sentence auditory presentation. We also used sentence-picture matching and grammaticality judgment as tasks in two self-paced listening studies with the same patients to measure on-line performance. In each task and presentation format, we presented sentences that tested the ability to assign and interpret three structural contrasts chosen to examine different basic syntactic operations: actives and passives, subject and object extracted relative clauses, and reflexive pronouns and matched sentences without these elements. We examined these behavioral data to determine patterns of impairment in individual patients and in groups of patients, using correlational analyses, factor analyses, and analyses of variance. The results showed that almost no individual patients had stable deficits referable to the ability to interpret individual syntactic structures, that a variety of structural features contributed to sentence processing complexity both on-line and off-line, that correct responses were associated with normal on-line and errors with abnormal performance, and that the major determinant of performance is a factor that affected performance on all sentence types. The results indicate that the major cause of aphasic impairments of syntactically based comprehension are intermittent reductions in the processing capacity available for syntactic, interpretive, and task-related operations.     

Cognitive and affective bases of opinion survey responses

Two experiments investigated the determinants of agreement with propositions that advocated social privileges for either people in general or specific social groups. Liking for the group to which a group-specific proposition referred had a contrast effect on agreement with a related general proposition that was considered immediately after it, but had a positive influence on agreement with a general proposition that did not occur until several items later. The latter effect was eliminated by instructing Ss to base their judgments on the consequences of the policies described in the propositions. Instructions to base judgments on affective reactions to the propositions produced contrast effects of group-specific propositions on judgments of general ones regardless of whether these items were separated or together in the questionnaire.