Positivism Philosophy With A Deductive Approach Psychology Essay

From Theory to Evidence: A Deductive Study on Violence Predictors Among Les Roches Students

This will not include subjective content and will be similar to a positivism philosophy where effort will be made to obtain data about observable reality and should a pattern be found in the data different theories can be formulated (Remenyi et al. (Saunders, 2009, p.144) Reproduced from 1998 For further information, please see:…Gilley & Johnson (2002, pp.39-40), also mention that the use of positivism leans towards a “cause and effect” relationship among the variables, and, the potential for identifying them. The reasoning behind this philosophy is the desire to find or establish a prospect decree (Fisher, 2010, p.32) model which supposedly depicts of Les Roches students probability to show violent behavior when exposed to predetermined amounts of variables/factors.
This is also consistent with research philosophy practiced by Homel and Clark (1994, p.1) who explain which factors (situational and managerial) were more projecting of violent behavior. Moreover, according to Saunders, Lewis and Thornhill (2012, p.135) empirical will be used through the positivism philosophy in order to carefully measure with quantitative data which factors contribute the most the the tendency of Les Roches students to exhibit violent behavior using “statistical analysis” as a tool to measure the results. However, the philosophy of positivism has received a great deal of criticism for its lack of “cultural sensitivity and historical specificity” (Calhoun,1992, in Mouton, 1996, pp. 200-201), as well as the fact that not much development is made in knowlege when the research is based on concepts that have already been established (Popper,1972, in Henn, Weinstein, & Foard, 2006, p.11). The greaeest point is that, positivism does not allow for appropriate interpreation and connotation of the findings from quantitative data, however this leads to an unrealistic generalisability (Swanson & Holton III, 2005, p. 18). However, Interpretivsm was not selected on the topic since the Research Question is not in the interest of describing nonrepresentational significance through observations but “expose the truth and facts of an issue as quantitatively specified relations among variables” (Swanson & Holton III, 2005,p.20).
Approach, Design and Time Frame
Using the “theory-then-research method” (Henn, Weinstein, & Foard, 2006, p.13) this research chose a deducive approach to test and generalise existing theories (Tharenou, Donohue, & Cooper, 2007, p.14), specifically related to factors influencing violent behaviour. Swanson and Holton III (2005, p.164) argue that deductivism relies upon a prior scale development process which mandates it employ a classification scheme for data collection. Many researches about this issue contents the same approach for example Hughes et al. s (2007,p.1) research, background information is already known, the decided deduction approach as they “explore variation in alcohol use and negative nightlife between young people who drank before going out or who do not consume alcohol until they arrive at bars and nightclubs”. One benefit of ordering from Top down, is that, if done correctly, it will help assure content validity in the end product (Swanson & Holton III, 2005, p. 165). Interestingly tough, Liang (1967, cited in Jill & Johnson, 2002,p.41) emphasizes the concern for the reason that it is based on scientific analysis and testing (Saunders, Lewis, & Thornhill, 2012, pp. 124-125), this will not take into account the natural science of the human behaviour in the ultimate analysis of the data, which may led to the fact of wrong generalization in the final answer and/or oversimplifying (Fisher et. al, 2010, p. 65) is caused by too many one’s tend to be thinking that the findings will be the absolute one there is, based on previous research and do not attempt to falsify it (Popper, 2002, cited in Fisher et. al, 2010, p.65). The goal of this research is to identify the cause and effect that independent variables (consumption, environment, management practice and cultural heritage) can have on the production of a dependent variable (violent behaviour), hence an explanatory research design was selected (Saunders, Lewis, & Thornhill, 2012, p.172). Due to the recognition of independed variables as the results of the primary data analysis, exploratory and descriptive research design methods were not utilized (Swanson & Holton III, 2005, p.117). The data collection snapshot strategy used was most aligned with a cross-sectional time frame, which suggests that it beeing at a particular time (Swanson & Holton III, 2005, p.80), given the time-sensitive nature of the study, findings (Saunders, Lewis, & Thornhill, 2012, p.190). In addition, the cross-sectional time frame is “cheaper and easier to organize” (Jupp, 2006, p.31) which means, since the study needed to be completed up to a certain time, longitudinal time framing was ineffective; it can take anywhere from months or years to decades (Jackson, 2009, pg. 327).
Unfortunately cross-sectional studies are said to be weak on “internal validity” (Tharenou, Donohue, & Cooper, 2007, p.20).
Research Strategy
Surveys were selected as the strategy of choice to gather the data needed to answer the research question. As driving this study Fisher (2010, p.69) reminds, surveys provide a relatively simple approach to examining the relationships between objects, helping us generate ideas about possible strong or weak associations or correlations between these variables.
According to Saunders, Lewis and Thornhill (2012, p. 173) Experiment’s and survey’s are mainly, and quite represent the two strategies used in quantitative data when it comes to research design, hence archivial research, case studies, ethnogrpahy, action research, grounded theory narrative inquiry were not appropriate for this study. The follow as to why survey strategy was chosen among other strategies relevant to this study is due to a survey’s ease of reach to a large population (Jackson, 2009, p.16), while the downside is the risk of the members of the sample group not being a fair cross section of the population, leading to possible false generalization. Nevertheless, Fisher (2010, p.69) notes that when enquiring on how to obtain a broad and representative overview of a situation the recommended approach is through surveys. Moreover, use of surveys is widely used in many studies related to this subject, like by Donath et al. (2011, p1) which applied use of survey’s across European High Schools to compare “how different migration backgrounds and rural vs. urban residence affect alcohol consumption and binge drinking in adolescents”.
Data Collection and Sampling
For secondary data, online scholar databases were searched and data was selected based on its relevance toward the current study. A lot of, research, were, published on similar issues, consumption (Hughes et al.,2007;Homel & Clark, 1994), environment (Livingston,2008; Macintyre & Homel,1997), management practices (Romain-Glassey et al., 2012; Roberts, 2007) and cultural heritage (McNulty and Bellair,2003; Donath, et al., 2011). The access to such secondary data allowed to develop and conduct solid and appropriate surveys that highly contributed to the construction of theoretical framework for this research (Saunders, Lewis and Thornhill, 2012, pp.304-307), as well as allowing to (Jill and Johnson, 2002, p.151) and supporting in the final comparison and explanation of the final results (Bryman and Bell, 2003, p.212). As Ghauri and Grønhaug (2005, quoted in Saunders, Lewis, & Thornhill, 2009, p.268) point out one way that secondary data collection helps is its relative facilitation in saving both time and money. Even so, one of the top concerns for the use of secondary data is whether definitions used in past research are indeed accurate or perhaps construed according to the researchers management making the differences of the analyses of data in fact inaccurate (Saunders, Lewis, & Thornhill, 2009, p.271), particularly in the case of qualitative data (Jupp, 2006, p.275).
The main data-collecting approach used in this study was mixed delivery and collection questionnaire and web-based questionnaire so that the data can be examined and stored easily (Swanson & Holton III, 2005, p.108). Data for most of the questionnaires was collected in writing on printed paper copies, randomly distributed to Les Roches Students who met the selection criteria sample for this study. Local authorities and venue owner were called and asked if they would agree to partake in the questionnaire and after the approval, the data was collected via structured interviews in accordance to the original questionnaire the Les Roches students had distributed. This shows how the use of actual data collection is important, as advised by Waters (2011, p.85), in that it is both recent and also reliable, their appropriateness to the study. However, the gathering of primary can be expensive and, in most cases, time consuming (Saunders, Lewis, & Thornhill, 2012, p. 306).
Survey questionnaires were employed nonetheless, due to their effectiveness (Bryman, 2004, cited in Tharenou, Donohue, & Cooper, 2007, p. 102) and prevalence in many studies regarding “attitude, values, preferences, intentions and personalities” (Moorman & Podsakoff,1992 cited in Tharenou, Donohue, & Cooper, 2007, p. 102). A web-based software was used text generating the questionnaires so that clear-cut questions are produced to guarantee reliability and internal validity (Saunders, Lewis, & Thornhill, 2012, pp. 428).
For instance, the questionnaire itself was presented within four A4 pages, since research indicates that lengthy questionnaires are discarded and have less response rates (Edwards et al. Saunders, Lewis, & Thornhill, 2012, pp. 445-446). The Les Roches students questionnaire was divided in three sections: The first part was focused in the demographic information of the surveyed, while the second part, with the use of a 5-point likert scale, measured the perceived influence assigned to different violent-triggering factors and the effectiveness of prevention methods. Finally the third section provided multiple choice responses that ask about the student current “party” habits,.
Fisher et al., (2010, p.227) noted that one of the most common response methods we choose when we need to know respondents opinions and attitudes are likert scales. Also according to Cook et al. According to (1981, cited in Swanson & Holton III, 2005, p.166) likert type scales are the most frequently used scaling forms in this kind of research. Furthermore, multiple choice questions typically give the people responding, two to five alternatives to answer a question and they have to select one (Fisher et al, 2010, p. 212). In questionnaires, there can be variation with the type of questions inquired, to do so would hinder the impacts of biased or skewed results (Henn, Weinstein & Foard 2006, pp.139).
The rationale for asking in these topics is because each of the information asked in the questionnaires has immediate significance on every independent variable (consumption, environmental, management concerns and cultural background), which all have an impact on the dependant variable (Powell, 1998, pp.2-3), which ends up being violent behavior.
Considering the Hughes (2006, p.1) presented a positive correlation between (the younger the) age of individuals in high number of violent episodes produced by binge drinking demographics were asked in section one of the questionnaire. Demographic questions were asked (see Table 1: Error: Reference source not found: Error: Reference source not found, Error: Reference source not found, Error: Reference source not found) so that author could acquire primary data from participants for comparison to prior literature (including Hughes studies). With regards to factors influencing violent behavior, and referencing previously reviewed studies by other prominent authors where consumption, environmental, management, cultural heritages were noted as leading factors (Homel and Clark, 1997; Macintyre and Homel, 1997; Shaw and McKay,1969), respondents were required to range their opinion in relation to the aforementioned factors in the completed questionnaire in section two. Also, it was asked the evaluability of prevention methods, since one of the works showed that these methods have efficacy in reducing violent behaviors, though it highlighted that such changes will vary according to the context and type of clientele (Miller et al., 2011; Mazerolle et al., 2012; Rossow & Nordstrom, 2011). Lastly, in the last section respondents were questioned about their current habits when “going out”. This was in reference to the vast amount of research which links a positive correlation of individuals who frequently go out and consume alcohol to high level of altercation (Homel & Clark, 1994; Romain-Glassey et al, 2012). To determine the main causes for violent behavior, the venue owner and authorities were also asked if the brawls seen inside their venues had been primarily instigated by Les Roches students or by locals.
The purpose of this study is, in due course to analyze what factors contribute towards violent behaviour in terms of Les Roches students, whether this is a student’s age, nationality or drinking habits, chronic or other.
As for the technique that was used for sampling of this study, probability sampling with simple random system was the optimum sampling technique due to “stratifying and clustering the sample to increase the precision” (Henn et al., 2006, p.129). The sample selected was stratified upon students thathad already been “engaged” for at least a semester in the Les Roches campus, therefore first year students (BBA1) were not included in the study research as they were not acquainted yet with the “The Les Roches Way of Life”. At least one student of each year-group filled the questionnaire, and was approached directly or indirectly through the teachers who were contacted separately to help in distributing the questionnaire, that way the concept of stratification and clustering was improved.
Probability sampling provides a “statistical framework that the sample drawn is representative of the population as a whole” (Swanson & Holton III, 2005, p.101). Also, Fink (1995, cited in Swanson & Holton III, 2005, pp.101-102) adds that “each individual studied will have a known, nonzero probability of being selected for the sample.” Additionally, this method is commonly associated with survey strategies (Saunders, Lewis, & Thornhill, 2012, p. 262). This method is preferred in most cases because it is superior to the non-probability technique: according to Tharenou, Donohue, & Cooper (2007, p.53) probability sampling is superior in that it strengthens “the external validity of a study”. When the sampling is made to separate the elements of the research population from the sampling frame with the best possible systematic decomposition, the possibility of making systematic discrepancies greatly expands, which becomes an even greater disadvantage and difficulty in the use of this technique (Jill & Johnson, 2002). In addition, Jupp (2006, p.197) adds that it can take time and turn out to be costlier to use. The reason of not using a non-probability technique as Henn, Weinstein, & Foard (2006, p.129) state, can only be used for those characteristics of a population that are known.
To calculate the right number needed for these students to be qualified to represent the whole population, an arithmetic table was used that was considering the total number of the population for an acceptable margin of error and the confidence level (Jill & Johnson, 2002, p.102), which brought the conclusion that a total of 250 surveys were needed.
In the case of local authorities and venue owner, however, snowballing a nonprobablistic purposive technique method of selection was adopted because there was a logistical and direct need to procure “a biased group for the purpose of screening” (Cooper & Schindler, 2003, cited in Tharenou, Donohue, & Cooper, 2007, p.56) to serve the reliability testimonium method of triangulation. Purposive sampling also stands out in being conducted in cases where the subjects of the sample group are for the most part directly related to the topic of issue (Swanson & Holton III, 2005, p.52). Neuman (2005 cited in Saunders, Lewis, & Thornhill, 2012, p. 287) highlites the technique as being used when the sample is relatevely small, which is evidentin this study and as well when the individuals in the sample are known toprovide the right kind of information needed. However, since this process is subjective to individuals bias, the drawback is clear depending of the willingness of one or more subjects of the sample to “make a particular point” and generalizing the solutions to its position (Swanson & Holton III, 2005, p.52)
Data Analysis
So far, the research procedure selected• Follow a deductive approach of quantitative data collection• Emphasis on finding casual relationship The analysis was then attempted through the numerical and statistical analysis of data (Saunders, Lewis, & Thornhill, 2009, p.175), where data was analysed through the Statistical Package for the Social Sciences Version 21 (SPSS) system of statistical analysis. It will also use simple linear regression analysis to explore the relationship between all independent and a dependent (Jupp, 2006, p. 65), to finish off with an evaluation of the model (see Error: Reference source not found), although SPSS is particularly efficient with statistical analysis (Ghauri & Grønhaug, 2010, p. 164), its plasticity is currently met with the challenge of finding an appropriate matching of variables and loss of information that was not able to satisfy the matching criteria thus risking losing out on statistical efficacy (Kleinbaum & Klein, 2010, p. 394). However, Lee & Lings (2008, p.321) show that this is the most popular quantitative analysis software used in research.
In analyzing the data, two forms of statistics were used; descriptive statistics which arrange the data collected in tables and graphs that represent percentages and values that will assist in the assessing and analyzing of the data (see Error: Reference source not found chapter), albeit will not represent as much information from what was generated as inferential statistics (Jackson, 2009, p. 109). On the other hand, inferential statistics, which will be engaged in generating generalized predictions of the whole population (Jupp, 2006, p. 147) based on an examination of a sample of the population from observations and analyses (Fisher, Buglear, Lowry, Mutch & Tansley, 2010, p.239). In addition, this data analysis process allows for researchers “to move beyond the bounds of their study sample” and describe likely new deductions for the entire population based on the sample observed (Marczyk, DeMatteo, & Festinger, 2005, p.224). Although this may be true, Mouton (1996:139) points out a critical concept to keep in mind that, the sample must be representative of the population to which it is generalised1. Thus, the number of respondents (N=250) that filled the questionnaire Estimated Sample Based on the confidence of 95% and margin of error of 95% shown that the sample examined can be generalized to the entire population.
In this case we are selecting inferential statistics which there are various statistical tests which can be parametic tests or nonparametic tests (Marczyk, DeMatteo, & Festinger, 2005, p.227). The parametic tests (z test, Pearson r and t test) are the tests that an assumption of the population is made by the researcher, using the mean, standard deviation and normal distribution with the use of interval (0-100) or ratio data (Jackson, 2009, pp.171-172). On the other hand, the nonparametric test (chi-square, Spearman’s Rho and goodness-of-fit tests), does not deal with population parameters such as mean or standard deviation or normal distribution, and involves ordinal or nominal data (Lee & Lings, 2008, p.336). Sig. p is equal or less than value `0.05`
Pearson r has a minimum value of -1 and a maximum value of r(+1), the sign in front of the number indicates whether the correlation is negative (one variable goes up, the other goes down) or positive (one variable goes up, the other goes up too) (Pallant, 2007, p.126). Perfect correlation is -1, or +1, that is, perfect predictability of one variable when the value of another variable is known (Tabachnick & Fidell, 2007, pp. 56). Cohen (1988, pp. 79-81, cited in Pallant, 2007, p.132) put forward the following “rule of thumb” in order to interpret the values between 0 and 1.
r = 0 –.29 weak relationship
r =. 30 –. 49 moderate relationship
r =. 50 – 1.0 strong relationship
The data analysis used a parametric testing of Pearson r as the data was interval or ratio. Moreover, according to generalisability, we chose the Pearson r because it is more accurate, also, choosing other testing methods would not have changed markedly the final statement since the high No of respondents (N=250) render an accurate generalisation for the population of Les Roches independent from the parametric test that we chose.
Ethics
Like Jupp (2006, p.96) argues, research insists that it is for the good of all to justify itself, but even so, there are ethical issue when conducting the research in feelings between this aim and the anxiety that it provokes in those who execute it. In this study the groups directly benefit from the research are Les Roches students the local government and the venue owner. Tharenou, Donohue, & Cooper (2007, p.317) argue that researchers should be aware of any potential ethical issues that could arise as their project develops before they even attempt to conduct a research project. In fact, Savin-Baden & Howell Major (2010, p.20) state that “Many Universities now require students to first seek ‘ethical approval’ of the research they plan to conduct before they start their practical analysis of their research”.
Many ethical principles must therefore be acknowledged, especially if humans are research subjects, such as being “respectful to the individuals”, granting them autonomy, as well as certain protection to individuals without autonomy (Tharenou, Donohue, & Cooper, 2007, p.317; Marczyk, DeMatteo, & Festinger, 2005, p.255) along with protecting confidentiality and anonymity (Fisher, Buglear, Lowry, Mutch, & Tansley, 2010, p.93). Questionnaires contained relevant detail regarding the topic of research on the first page, and the principle of confidentiality and anonymity was also explained in the questionnaires. Besides, the students received more information about the subject when the questionnaires were provided. Formal e-mail requests were made to local authorities and the venue owner for participation in the study before conducting the semi-structure interviews. During the semi-strucured interviews, the owner of the venue and the police authorities were ensured that their identity is going to remain anonymous in the research, so that their personal and institutional privacy would be protected (Swanson & Holton III, 2005, p. 109).
Furthermore, the research needs to be conducted in a way “that avoids all unnecessary physical and mental distress and/or harm” (Henn, Weinstein, & Foard, 2006, p.71). Due to the sensitive subject matter of this study, an experimental research strategy was not employed, nor was a case study analyzed so as not to cause physical harm to the researcher. This explains why the less harmful approach to take for this study were surveys, that only depended on participants rememberance of past points of violence instead of recent ones. Since we know that culture was a major parameter involved in this analysis, it became necessary to ensure that we did not lean towards any discrimination or biasness towards one particular response. This included the omission of drug use and its contribution to violent conduct which numerous researchers (Powell, 2011, p.1; Goldstein, 1985, p.1; Homel & Clark, 1994, p.24) have noted as relevant to aggression, based upon the pilot test and supervisory suggestion. Moreover, there is also an ethical side of this issue related to “Zero Tolerance” already laid out in the Les Roches International School of Hotel Management policy.
Then, the study must adhere to the principle of honesty, with regards to its participants, must be made in good faith and not discrimination in the selection and use of respondents (Thorpe & Holt, 2008. p.30). The sampling method used was simple random sampling among Les Roches students who were a part of campus life for over one semester, as previously stated. Other than that, there was no discrimination based on participants nationality, sex or age.
Limitations and Credibility
Limitation
There are limitations in the study which have to kept in mind while thinking about data in the research. According to Marczyk, DeMatteo, & Festinger (2005, p. 35), while research serves a key role in illustrating the relationship of two things, “the primary limitation of a study is that a correlation (the relationship between two things) does not imply causality.”
In this research, while collecting data from students of Les Roches, the major issue is that either the students might be providing false data with respect to their ranking of factors dominating violent behaviour, as that could be confusing or they may want to make themselves look good, a sense of anxiety to complete the form, they might involuntarily miss answers, then does not give the research proper results. The reliability of the study was further reinforced using a pilot test Cronbach’s alpha score of . Surprising result of this study was internal reliability within a range of one of the pillars in accordance to Bryman & Bel (2007, p.164) implicating very good level, indeed, as it passes minimum value according to Nunnally (1978,cited in Pallant, 2007, p.20) this value should exceed 0,7 minimum.
The study was cross-sectional, so it obviously will not pick up changes over time. Another thing to keep in mind about this fact is that this study will be carried out in winter at the school semester and therefore there will not be many students who go out to hang out in sites and they are exposed to much violence, since the majority of students   *will not leave their dorms due to the severe cold weather for such their opinions will be less expensive than if a sample of Les Roches Int. summer population. This research would gather data from School of Hotel Management. A further limitation may be implicated in the subjectivity between inadvertent and intentional contacts between patrons. It is also worth noticing that violence can be a lot more prevalent in other places than others due to the type of clientele it attracts, so a special theme night may cause violence to happen at a much higher frequency.
Regarding representativeness, as a non-probability method of convenience sampling was taken(is done) for the venue owner and local authorities, the data received from the interviewees will be generalized to the limited amount of individuals that were interviewed. Given this, the results must be interpreted with caution and the final analysis considers the limitations towards the generalisability of results toward licensed venues or authority on a national or global level. Also with regards to the Les Roches students it may not represent the population as the simple random sampling method was a threat, as a certain course of nationality may dominate the participants (Jupp, 2006, p. 171), which is often the case with studies involving human beings (Jill & Johnson, 2002, p.60), resulting in a bimodal distribution (Hatcher, 2003, p. 198).
Validity and Reliability
As such, in order to strengthen the validity of the data obtained, triangulation was used as it pushed the study to derive the information from varied viewpoint, discussing varied perspectives on the topic selected and thus providing the obtained final analysis to be scrutinized as well as giving the opportunity to conduct an overall detailed analysis (Henn, Weinstein, & Foard, 2006, pp.20-12). The disadvantage of this method is that it takes time and experience in both qualitative and quantitative method (Swanson & Holton III, 2005, p. 319), adding to the reliability of the study. Moreover, as it is a time sensitive paper it can be hard to analyze as it creates a lot of statistical data which leads to large observable correlations (Henn, Weinstein & Foard, 2006, p. 215). In addition, the data was considered statistically significant based on the parametric testing of Pearson r as detailed in section 1.5 Data Analysis.