Evaluating Farmers’ Access to Agricultural Information: Evidence from Semi-Arid Region of Rajasthan State, India

Comment

Response

Reviewer 1

Summary: The authors examine   whether the availability/use of ICTs improves access to information among   rural farmers in Rajasthan. They differentiate between information on   agricultural production and information on agricultural marketing in the   analysis —and between different categories of information on production and   marketing. The authors find that ICTs users make greater use of the different   sources of information and thus recommend improving farmers´ skills in the   use of ICTs in order to increase their use of information.

While I find that   the role of ICTs on farmers´ access to information can be of key importance   for policy making, I think the paper in its current version can be improved   in several aspects:

Thank you very much   for your encouraging comments.

Broad   comments:

I miss a description   of the economic theory behind the article, as well as of the background of   the study —area of study and other features which may affect the results of   the study and their implications for the analysis. I would suggest the   authors providing some information on both.

I think more   information about the survey and the data should be provided —more detail on   the variables collected in the survey, some information about the survey´s   design, whether it is nationally representative or to what extent it is   representative, and the source of the relevant auxiliary information which   was “acquired from reliable sources” (line 105).

I would suggest the   authors using a binary variable instead of X1 and X2 in the regression   analysis.

I would also suggest   considering the possibility of using additional control variables in the   regression in order to gain additional insight into the study.

Thank you very much   for your critical comment.

Please we have now   specified the economic rationale behind the article as follows:

The economic   rationale for the farmers’ access to information is to enable them to the   manage risks and uncertainties regarding production and marketing of their   produce. The better the farmers manage these risks and uncertainties the more   profitable their businesses become. ICT facilitates awareness and access to   market information among the farmers [4]. There are more than 200 ICT   development agencies in different stages of implementation in India e.g.,   Bhoomi, Drishtee. These agencies provide information relating to for   instance, climate reports, and marketing information e.g., Krishi Vigyan   Kendras/Farm Science Centers at Ahmednagar, Baramati) [11].  

(Please see page 3   lines 82 – 88).

We have also   provided further information on the area of study. (Please page 4-7).

Please, we have now   provided detailed information about the survey including the sampling   procedure, population and sample size. (Please see page 8).  

The survey was   carried out in three districts in Rajasthan State in India. Technically, the   data and results obtained in this study may not be fully representative of   the population of India. Therefore, we have specified this as the key   limitation of the study and provided suggestions for future research in the   conclusions (see Section 4, page 18 – 19).

Also, we have now   referenced all sources of reliable information that we have obtained for this  study.

Thank you very much for this comment. Please the data obtained for this study is not binary. The  five-point Likert questionnaire was used to obtain the data. Thereafter, the values were averaged using the weighted average as specified on page 8.

In this study, the additional explanatory variables were not used for the regression analysis. Previous studies have examined these relationships as we have noted on page 3   lines    90 – 95. The crux of this   study is to examine whether information from ICT sources influence the   farmers’ overall access to production and marketing information for their   businesses.

Specific   comments

Response

Point   1:   I suggest not using so many subsections in Section 2.


Thank you very much.   Please we have now reduced the number of subsections in Section 2 and other   parts of the manuscript as suggested.

Point   2:   I think the specification of the WAI should be better explained (lines   131-145). Were the “different levels of access” used in its calculation   collected in the survey? Please indicate whether each category of information   used in the analysis is assigned one of these levels of access in the survey.   Otherwise provide information on how the variables are computed.

Thank you very much.   Yes, the different levels of access used in its calculation were collected   during the survey.

The WAI is used   calculate the level of access of individual production and marketing   Information separately.

Thus, access to production   and marketing information were evaluated based on the data collected by using   the weighted average index (WAI) specification on page 8 line 228 – 235).

Point   3:   Is the dependant variable Y used in the regression the WAI? If so, the   authors should say so.

Thank you very much.   Please No.

The WAI was calculated   for each type of production and marketing Information.

Thereafter, the averages   of all types of production information were used to derive the production   mean for the first multiple regression analysis.

Similarly, the averages   of all types of marketing information were used to derive the marketing   information mean for second multiple regression analysis.

Please we have now indicated this as a footnote 3 on page 9.

Point   4:   Line 153: do you refer by any chance to household heads?

Thank you very much.   Yes, household head.  These farmers are main source of income and take   major decisions for their households.

Point   5:   Lines 153-171: I would suggest including an additional table which contains   the descriptive statistics. It would improve the paper.

Thank you very much.   The table of descriptive statistics has now been added to provide further   insight. (Please see section 3.1, Table 2, page 9 – 11).

Point   6:   Outline how the “Production Information (mean)” and the “Marketing   Information (mean)” are defined (from line 177 on).

Thank you very much.  

Production mean is   the average of all types of production information accessed.

Marketing mean is   the average of all types of marketing information accessed.

Please refer to our   response to point 3 above.

Point   7:  I would also suggest mentioning the fact   that the authors are differentiating between information on production and   information on marketing —and the categories or items they are considering   within each of them— earlier in the paper.

Thank you very much.   Please This information is has now been added to the presentation of the Conceptual   Framework of Research.

(Please see page   4). 

Point   8:   I would suggest including the fact that the authors are undertaking the   analysis on multiple comparisons too in the section on material and methods   (the results of the analysis are included in Tables 2 and 5) and giving an   explanation for its inclusion in the analysis.

Thank you very much.   Please the methods of analysis have now been properly presented in the   revised manuscript as suggested.

(Please see page 7 –   9).

Point   9:   Figures 1, 2 and 3: Could you please explain what the equivalent rank is?

Thank you very much.    The equivalent rank is   the percentage equivalent of ICT source accessed out of Total (ICT+ Non-ICT).    

Point   10:   Please number the equations.

Thank you very much.   Please we now sequentially numbered all the equations in the manuscript.

Point   1: The authors   should first provide a theoretical background/framework of the study than can   help them identify the objectives of the paper. The study of the   socioeconomic characteristics of the respondents cannot be a scientific   objective; this is a tool for the analysis. The authors should outline the   role of farm training/ farm education on farm economic performance/ labour   productivity (there are a lot of references in the international literature,   e.g., Exploring the labour productivity of agricultural systems across   European regions: A multilevel approach) and then explore the role of ICTs on   achieving this.

Thank you very much.

The theoretical background and conceptual   framework of the study have now been provided.  (Please see page 3 and 4).

The socioeconomic   characteristics of respondents have been removed from the list of objectives   as suggested. (Please see page 4).

Also, an exposition of the   influence of education and other socio-economic factors as well as farm level   factors on ICT access have been provided in the revised manuscript.(Please see page 3).

Point   2: The authors   should enhance their multiple regression model. All the variables derived   from the socioeconomic survey (e.g., age, education, perceptions etc.) should   be included as explanatory variables (control variables) in the model to test   their impact. Also instead of using X1 and X2 for access/not access to ICT,   the authors could think of using a binary variable (1: access, 0: otherwise)   avoiding thus multicollinearity issues

Thank you very much for this   comment. Please the data obtained for this study is not binary. The   five-point Likert questionnaire was used to obtain the data. Thereafter, the   values were averaged using the weighted average as specified on page 8.

In this study, the additional   explanatory variables were not used for the regression analysis. Previous   studies have examined these relationships as we have noted on page 3 lines 90   – 95. The crux of this study is to examine whether information from ICT   sources influence the farmers’ overall access to production and marketing   information for their businesses.

Point   3: The authors   should provide more information about the area of the study and mainly the   agricultural sector: contribution of agriculture to total GDP and employment,   prospects, dynamics.

Thank you very much. Please we   have now provided more information about area of study.

(Please see section 3.1).

Point   4: The authors   should provide more details about the design of the survey. What is the farm   population size of the case study and what the 133 respondents represent;   synthesis of livestock breeders and farmers etc.

Thank you very much. Please,   further information on the design of survey, farm population size and   information about 133 respondents have now been provided.

(Please see page 8).

Point   5: The article   is missing policy recommendations.

Thank you very much. Please,   the policy recommendation emanating from the conclusions of the study has now   been provided. (Please see section 4, page 19-20).

Even though the paper has improved, I still have a few concerns.

Thank you very much.

 I would suggest the authors   combining both maps into one (Figures 3 and 4).

Thank you very much. Please we have now combined the two figures into   one as suggested (Please see Figure 3).

In the description of variables section: Why are   equations 1 and 2 divided by 2? Besides, I am not sure I would consider   information obtained through own knowledge and experience as “information”.

Thank you very much for this comment.

The reason is as follows:

We have used equal number of ICT and Non- ICT sources.   The % ICT source from which the information was accessed is calculated based   on Total ICT sources. Similarly, the % Non-ICT source from which the   information was accessed is calculated based on Total Non-ICT sources.

Therefore, to calculate % ICT access out of Total   ICT + Non- ICT access, it was divided by 2 as specified in equations (1) and   (2).

Please we have now included this explanation as a   footnote in the manuscript (i.e. footnote 3).

Also, we absolutely agree with you that information   obtained through own knowledge and experience cannot be considered as   information. Therefore, we have deleted equation (3) from the manuscript,   more so, that this is not part of the discussion in the manuscript.

I would suggest including Table 2 in the “sampling and data” section   or in a new different section (descriptive statistics are not results). I   think it could be a good idea to include statistics for both users and   non-users in two different columns. It would give additional insights into   the study.

Thank you very much for this suggestion. Please   we have now included the socio-economic characteristics of the farmers as   part of section 2 of the manuscript as suggested and categorized them into   farmers who solicited information from ICT sources and Non-ICT sources.

In the original paper three groups of ICT users   were considered (ICT users/non-ICT users/moderate users). In the reviewed   manuscript the authors consider two only. I assume moderate users have been   included in the “ICT users” group. The Games-Howell post-hoc test does not   longer make sense. It is used to confirm which means (between the different   groups) are significantly different from each other. The ANOVA already   confirms there are differences between the two groups analysed. If the   authors still want to include the Games-Howell test, I would suggest removing   the “moderate” category.

Thank you very much for this comment.

Please we have added the moderate ICT users to   the “ICT users” group, and have removed the moderate category from the   discussion.

The moderate users were only 3 in our sample;   therefore, we thought it would be proper to add them to the ICT users group.

Consequently the presentation in this revised   manuscript is based on ICT users and Non-ICT users.

Consequently, we have now replaced the   Games-Howell post-hoc test with the Student’s t-test of the differences in access to different types of information   between ICT users and Non-ICT users (Please see Table 4 and 7).

As for the regression, I would still suggest using some additional   explanatory variables from the survey (if it is possible).

Thank you very much.

Please we have now included socio-economic and  

some farm-level variables in the regression (Table 5

and 8). The gender variable is significant in explaining farmers’ access to production and marketing

information, and the corresponding implication has

been provided.

The authors have addressed most of the comments

Thank you very much.